Atmospheric pollutant accurate traceability analysis method and system based on artificial intelligence
Through artificial intelligence, analyzing atmospheric component data and wind field maps, building a traceability path, combining historical records and enhanced learning samples, the problem of traceability errors in atmospheric pollutants is solved, the traceability efficiency and accuracy are improved, and the reliability of air pollution control is realized. stability.
Patent Information
- Application Number
- CN202411033054.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-30
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-07-30
AI Technical Summary
The prior art has errors in the traceability of atmospheric pollutants, resulting in errors in the positioning of pollution sources and affecting traceability efficiency and accuracy.
The precise traceability analysis method of atmospheric pollutants based on artificial intelligence is adopted to construct traceability paths by obtaining atmospheric component data and real-time wind field maps, and combine historical traceability records and enhanced learning samples to conduct pollution source positioning early warning and verification.
It improves the accuracy and efficiency of traceability of atmospheric pollutants, promptly controls air pollution, and improves the reliability and stability of governance.
Smart Images

Figure CN118937586A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the field of computer technology, and in particular to an artificial intelligence-based precise source tracing analysis method and system for atmospheric pollutants. Background Art
[0002] At present, with the rapid development of economy and industry, the emission rate of air pollution is getting higher and higher. Atmospheric environmental pollution has become the most serious environmental safety problem, which has a great impact on people's quality of life. In order to timely discover and control air pollution, it is necessary to accurately and efficiently trace the source of air pollutants. To this end, detection stations are usually arranged at various locations in the detection area to monitor the atmospheric composition data in real time. Then, the background analysts observe the changes in the atmospheric composition data to determine whether atmospheric pollutants appear. When atmospheric pollutants appear, go to the relevant sites in time to search for the source of pollution.
[0003] However, since atmospheric pollutants will drift with the wind direction, the characteristics of atmospheric pollutants represented by atmospheric composition data are often more obvious than the actual location of the pollution source at the downwind location of the pollution source. The method of simply locating the emission location of atmospheric pollutants based on atmospheric composition data and then tracing the source of atmospheric pollutants has large errors, which can easily lead to incorrect positioning of the pollution source and affect the efficiency and accuracy of tracing the source of atmospheric pollutants. Summary of the invention
[0004] The embodiments of the present application provide an artificial intelligence-based precise source tracing analysis method and system for atmospheric pollutants, which can improve the accuracy and efficiency of tracing atmospheric pollutants and solve the problem of errors in tracing atmospheric pollutants.
[0005] In a first aspect, the present application provides an artificial intelligence-based precise source tracing analysis method for atmospheric pollutants, including:
[0006] Acquire atmospheric component data information of a target detection area, input the atmospheric component data information into a pre-constructed atmospheric pollutant characteristic detection model, output target atmospheric pollutants and corresponding target associated industry information of the target detection area based on the atmospheric pollutant characteristic detection model, the atmospheric pollutant characteristic detection model constructs training samples based on atmospheric component characteristic data of different atmospheric pollutants and corresponding associated industry information in advance, and performs model training based on the training samples;
[0007] Determine the target index data to be detected based on the target air pollutants in the target detection area, and collect the target index data detected by the collection nodes at each designated detection position in the target detection area;
[0008] Acquire a real-time wind field map of the target detection area, mark the target index data on the real-time wind field map based on the designated detection position, and determine distribution information of the target index data on the real-time wind field map;
[0009] A traceability path for the target detection area is constructed based on the distribution information, a target location point matching the target-associated industry information is searched within a set range of the traceability path, and a pollution source location warning for the target atmospheric pollutant is performed based on the target location point.
[0010] Furthermore, the step of constructing a traceability path of the target detection area based on the distribution information includes:
[0011] The distribution path of the target atmospheric pollutant is constructed according to the distribution information, and the section on the distribution path that matches the wind direction of the real-time wind field map is determined as the tracing path of the target detection area.
[0012] Furthermore, the step of constructing a traceability path of the target detection area based on the distribution information includes:
[0013] A pollution concentration gradient change path of the target atmospheric pollutant is constructed based on the distribution information. When the pollution concentration gradient change path matches the wind direction of the real-time wind field map, the pollution concentration gradient change path is determined as the tracing path of the target detection area.
[0014] Further, after the pollution source positioning and early warning of the target air pollutant is performed based on the target location point, it also includes:
[0015] A detection path is constructed according to the wind direction data of the target location point and the real-time wind field map, and unmanned equipment is driven to collect pollution concentration change information corresponding to the target atmospheric pollutants in the detection path, and the target location point is traced and located based on the pollution concentration change information.
[0016] Furthermore, after driving the unmanned equipment to collect the pollution concentration change information of the target air pollutant corresponding to the detection path, it also includes:
[0017] Based on the pollution concentration change information, it is detected that the pollution concentration difference at the designated position point of the detection path reaches a set threshold, and a pollution source positioning warning of the target atmospheric pollutant is performed based on the designated position point.
[0018] Further, after determining the distribution information of the target index data on the real-time wind field map, the method further includes:
[0019] Query the historical atmospheric pollutant source tracing record database, determine the historical atmospheric pollutant source tracing records that match the real-time wind field map and the distribution information, and extract the historical pollution sources of the historical atmospheric pollutant source tracing records to serve as the tracing reference location points for the current target atmospheric pollutants.
[0020] Furthermore, after collecting the target indicator data detected by the collection nodes at each designated detection position in the target detection area, the method further includes:
[0021] A reinforcement learning sample is constructed based on the target indicator data and the target air pollutant, and the air pollutant feature detection model is trained based on the reinforcement learning sample.
[0022] In a second aspect, the embodiment of the present application provides an artificial intelligence-based precise source tracing analysis system for atmospheric pollutants, including:
[0023] A model detection module is used to obtain atmospheric component data information of a target detection area, input the atmospheric component data information into a pre-constructed atmospheric pollutant characteristic detection model, and output target atmospheric pollutants and corresponding target associated industry information of the target detection area based on the atmospheric pollutant characteristic detection model. The atmospheric pollutant characteristic detection model constructs training samples based on atmospheric component characteristic data of different atmospheric pollutants and corresponding associated industry information in advance, and performs model training based on the training samples;
[0024] An index collection module, used to determine target index data to be detected based on the target atmospheric pollutants in the target detection area, and collect the target index data detected by the collection nodes at each designated detection position in the target detection area;
[0025] A distribution analysis module, used to obtain a real-time wind field map of the target detection area, mark the target index data on the real-time wind field map based on the designated detection position, and determine the distribution information of the target index data on the real-time wind field map;
[0026] The tracing module constructs a tracing path for the target detection area based on the distribution information, queries for a target location point that matches the target-related industry information within a set range of the tracing path, and performs pollution source location warning for the target atmospheric pollutants based on the target location point.
[0027] In a third aspect, an embodiment of the present application provides an electronic device, including:
[0028] memory and one or more processors;
[0029] The memory is used to store one or more programs;
[0030] When the one or more programs are executed by the one or more processors, the one or more processors implement the artificial intelligence-based precise source tracing analysis method for atmospheric pollutants as described in the first aspect.
[0031] In a fourth aspect, an embodiment of the present application provides a storage medium comprising computer executable instructions, which, when executed by a computer processor, are used to execute the artificial intelligence-based precise source tracing analysis method for atmospheric pollutants as described in the first aspect.
[0032] The embodiment of the present application obtains atmospheric component data information of a target detection area, inputs the atmospheric component data information into a pre-constructed atmospheric pollutant characteristic detection model, outputs target atmospheric pollutants and corresponding target-related industry information of the target detection area based on the atmospheric pollutant characteristic detection model, the atmospheric pollutant characteristic detection model pre-constructs training samples based on atmospheric component characteristic data of different atmospheric pollutants and corresponding related industry information, and performs model training based on the training samples; determines target indicator data to be detected based on the target atmospheric pollutants in the target detection area, and collects the target indicator data detected by the collection nodes at each designated detection position of the target detection area; obtains a real-time wind field map of the target detection area, marks the target indicator data on the real-time wind field map based on the designated detection position, and determines the distribution information of the target indicator data on the real-time wind field map; constructs a traceability path of the target detection area based on the distribution information, queries the target location point that matches the target-related industry information within the set range of the traceability path, and performs pollution source location warning of the target atmospheric pollutants based on the target location point. By adopting the above-mentioned technical means, the target atmospheric pollutants can be accurately detected through the atmospheric pollutant characteristic detection model, and then the target atmospheric pollutants can be accurately traced in combination with the real-time wind field. This can improve the efficiency and accuracy of atmospheric pollutant tracing, carry out atmospheric pollution control in a timely manner, and improve the reliability and stability of atmospheric pollution control. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 This is a flow chart of an artificial intelligence-based precise source tracing analysis method for atmospheric pollutants provided in Example 1 of the present application;
[0034] Figure 2 This is a schematic diagram of the interaction between the collection node and the tracing server in Example 1 of the present application;
[0035] Figure 3 This is a flow chart for determining the traceability reference position point in the first embodiment of the present application;
[0036] Figure 4 It is a schematic diagram of the wind field direction and the distribution position of the target atmospheric pollutants in the first embodiment of the present application;
[0037] Figure 5 This is a structural diagram of an artificial intelligence-based precise source tracing analysis system for atmospheric pollutants provided in Example 2 of the present application;
[0038] Figure 6 It is a structural diagram of an electronic device provided in Example 3 of the present application. DETAILED DESCRIPTION
[0039] In order to make the purpose, technical scheme and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for the convenience of description, only the part related to the present application but not all the contents are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow chart describes each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of each operation can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to a method, a function, a procedure, a subroutine, a subprogram, etc.
[0040] Embodiment 1:
[0041] Figure 1 A flowchart of an accurate source tracing analysis method for atmospheric pollutants based on artificial intelligence provided in Example 1 of the present application is given. The accurate source tracing analysis method for atmospheric pollutants based on artificial intelligence provided in this embodiment can be performed by an accurate source tracing analysis device for atmospheric pollutants based on artificial intelligence, which can be implemented by software and / or hardware. The accurate source tracing analysis device for atmospheric pollutants based on artificial intelligence can be composed of two or more physical entities, or it can be composed of one physical entity. Generally speaking, the accurate source tracing analysis device for atmospheric pollutants based on artificial intelligence can be a computing device such as an atmospheric pollution source tracing server or a computer.
[0042] The following description takes the source tracing analysis device as an example to implement the precise source tracing analysis method of atmospheric pollutants based on artificial intelligence. Figure 1 The precise source tracing analysis method of atmospheric pollutants based on artificial intelligence specifically includes:
[0043] S110. Obtain atmospheric component data information of the target detection area, input the atmospheric component data information into a pre-constructed atmospheric pollutant characteristic detection model, and output the target atmospheric pollutants and corresponding target associated industry information of the target detection area based on the atmospheric pollutant characteristic detection model. The atmospheric pollutant characteristic detection model pre-constructs training samples based on the atmospheric component characteristic data of different atmospheric pollutants and the corresponding associated industry information, and performs model training based on the training samples.
[0044] When tracing the source of atmospheric pollutants, the present application obtains atmospheric component data information of the traceable area to be detected, defines the traceable area to be detected as the target detection area, detects atmospheric pollutants through a pre-built atmospheric pollutant feature detection model, and determines the atmospheric pollutants present in the atmospheric environment of the target detection area.
[0045] Among them, by deploying atmospheric component monitoring equipment at the corresponding locations of the target detection area, these devices can measure and record various component data in the atmosphere, such as PM2.5, PM10, SO2, NOx, VOCs, etc. Atmospheric component data is collected from these monitoring devices in real time and transmitted to the traceability analysis device in real time. The traceability analysis device first cleans and preprocesses the collected atmospheric component data, including removing outliers, filling missing values, correcting measurement errors, etc., to ensure the quality and accuracy of the data. Then, atmospheric pollutants are detected through the pre-built atmospheric pollutant feature detection model.
[0046] Prior to this, a deep learning or machine learning-based atmospheric pollutant feature detection model was pre-designed. The model can receive atmospheric component data as input and output the identified atmospheric pollutant types and possible related industry information.
[0047] When training the atmospheric pollutant feature detection model, atmospheric component feature data containing different atmospheric pollutants are selected from the historical atmospheric pollutant source tracing database as input features. For each input sample, the corresponding atmospheric pollutant type and related industry information need to be marked. The related industry information can be determined through expert knowledge, industry surveys, or historical pollution event records. Then, the processed atmospheric component feature data and related industry information are combined into a training data set, in which each sample contains a set of atmospheric component feature data and its corresponding related industry information.
[0048] According to the complexity of the problem and the characteristics of the data, a suitable atmospheric pollutant feature detection model is designed. This model can be based on machine learning (such as decision trees, random forests, gradient boosting machines, etc.) or deep learning (such as convolutional neural networks, recurrent neural networks, etc.). This application does not impose fixed restrictions on the selection of specific model architectures, and will not be elaborated on here. Through model training, it is possible to capture the correlation between atmospheric component data and the types of atmospheric pollutants and related industries. The model is then trained using a training data set. During the training process, the model learns how to map the input atmospheric component feature data to the output types of atmospheric pollutants and related industry information. By continuously adjusting the parameters and structure of the model, the prediction error is minimized and the generalization ability of the model is improved.
[0049] After the model is implemented, when conducting atmospheric pollutant detection, the real-time or newly collected atmospheric component data information is input into the trained atmospheric pollutant characteristic detection model. Based on the input atmospheric component characteristic data, the model outputs the target atmospheric pollutant types and corresponding target related industry information in the target detection area. This information can serve as the basis for subsequent traceability analysis.
[0050] Optionally, with the emergence of new types of air pollutants and related industries, as well as the continuous development of monitoring equipment and technology, it is necessary to regularly update and maintain the air pollutant characteristic detection model to ensure its accuracy and effectiveness. By continuously collecting new atmospheric component data, and using these data to verify and calibrate the prediction results of the model, the stability and reliability of the model can be improved. In some embodiments, the atmospheric component data can also be fused with other relevant data (such as meteorological data, traffic data, industrial emission data, etc.) to further improve the accuracy and comprehensiveness of the atmospheric pollutant source analysis.
[0051] S120 , determining target indicator data to be detected based on target atmospheric pollutants in the target detection area, and collecting target indicator data detected by collection nodes at each designated detection position in the target detection area.
[0052] After determining the target air pollutants in the target detection area, it is necessary to clarify the target indicator data that need to be detected based on the target air pollutants and collect these data. These target indicator data are usually closely related to specific air pollutants and are an important basis for evaluating the concentration, distribution and change trend of pollutants.
[0053] First, conduct an in-depth analysis of the identified target atmospheric pollutants to understand their physical and chemical properties and their impact on the environment and human health. Based on the analysis results, determine the key indicators closely related to the target atmospheric pollutants. For example, for PM2.5, the key indicators may include its concentration, particle size distribution, chemical composition, etc.; for SO2, the key indicators may be its concentration and emission rate. In addition, the target indicator data can also be some related pollutant components, such as related sulfides that SO2 may convert in the atmosphere. According to relevant laws and regulations, environmental quality standards or scientific research needs, the detection standards and thresholds of the target indicator data are set accordingly. Then, within the target detection area, a collection instruction is issued to the pre-set numerous collection nodes within the target detection area to instruct the collection nodes to collect the target indicator data. The system will deploy appropriate monitoring equipment on each collection node, and the monitoring equipment integrates various types of sensors for real-time collection of target indicator data. These sensors can include gas analyzers, particulate samplers, meteorological stations, etc. The target indicator data corresponding to the designated detection location in the target detection area can be collected by numerous collection nodes to comprehensively and accurately reflect the distribution of the target atmospheric pollutants. The collected data is then transmitted to the traceability analysis equipment in real time to ensure the security and accessibility of the data, facilitating subsequent data analysis and processing.
[0054] For example, refer to Figure 2 In the target detection area A, the collection node 12 is pre-set at each designated detection position, and the collection node 12 communicates with the traceability analysis device 11. When receiving the collection instruction of the traceability analysis device 11, the target indicator data of the corresponding target detection position can be collected and summarized to the traceability analysis device 11 in real time.
[0055] Optionally, during the data collection process, strict data quality control measures are adopted, including equipment calibration, data verification, and outlier processing, to ensure the accuracy and reliability of the data. In addition, it is necessary to ensure that the collected data has a complete time series and spatial distribution to facilitate subsequent spatiotemporal analysis and pollution source tracking. For key target indicator data, a real-time monitoring and early warning system should be established to promptly issue an alarm when the data exceeds the preset threshold and take appropriate countermeasures. Through the above steps, the key indicator data of the target atmospheric pollutants in the target detection area can be efficiently collected, providing strong data support for the subsequent atmospheric pollutant source tracing analysis.
[0056] Optionally, after collecting the target indicator data detected by the collection nodes at each designated detection position in the target detection area, the method further includes:
[0057] Reinforcement learning samples are constructed based on target indicator data and target atmospheric pollutants, and an atmospheric pollutant feature detection model is trained based on the reinforcement learning samples.
[0058] After collecting the target indicator data detected by the collection nodes at each designated detection position in the target detection area, these data can be further used to optimize or update the atmospheric pollutant feature detection model. Among them, according to the collected target indicator data and the known target atmospheric pollutant information, a corresponding label is generated for each sample. These labels may include the type of atmospheric pollutants, concentration level, pollution source type, etc. In order to improve the generalization ability of the model, the original data can be enhanced. Data enhancement methods include but are not limited to noise addition, data transformation (such as rotation, scaling, translation), synthesis of new samples (generating new samples based on known samples but maintaining key features), etc. These enhanced data will constitute an enhanced learning sample set together with the original data. Then the enhanced learning sample set is screened to remove invalid or low-quality samples to ensure the accuracy and representativeness of the training data.
[0059] Afterwards, fine-tune or optimize the model parameters based on supervised learning (such as neural networks, decision trees, etc.) and reinforcement learning. The atmospheric pollutant feature detection model is trained by using a reinforcement learning sample set. During the training process, the model will learn how to extract features from the input target indicator data and predict the corresponding atmospheric pollutant types, concentrations and other information. By continuously adjusting the parameters and structure of the model, the prediction error can be minimized and the accuracy of the model can be improved. During or after the training process, the model can also be verified and tested using independent validation sets and test sets to evaluate the performance of the model and identify possible problems or deficiencies. Based on the results of verification and testing, the model is further optimized, such as adjusting model parameters, improving model structure, and using more complex feature extraction methods.
[0060] By constructing enhanced learning samples based on target indicator data and target atmospheric pollutants, and training more accurate and adaptable atmospheric pollutant feature detection models, the efficiency and accuracy of atmospheric pollutant tracing can be improved.
[0061] S130, obtaining a real-time wind field map of the target detection area, marking the target index data on the real-time wind field map based on the specified detection position, and determining the distribution information of the target index data on the real-time wind field map.
[0062] Furthermore, based on the determined target indicator data, the target indicator data is annotated on the real-time wind field map of the target detection area to observe the correlation between the target indicator data and the real-time wind field. In order to obtain the real-time wind field map of the target detection area and annotate the target indicator data on the map to determine the distribution information of the data, it is first necessary to obtain the real-time wind field map of the current target detection area from a data source such as a meteorological platform through an interface call or the like.
[0063] Optionally, when obtaining the real-time wind map of the target detection area, the real-time wind speed, wind direction and other data of the target detection area provided by a reliable meteorological data source can also be used to draw the real-time wind map. These data can come from a weather station, a meteorological satellite, a radar system or a professional meteorological service provider, etc. Real-time wind data is received through an API interface, a data transmission protocol (such as FTP, HTTP, etc.) or directly from a meteorological device. The received wind data is processed, including the conversion of wind speed and wind direction (such as converting from a vector form to a format required for graphical representation), interpolation processing (to fill in data gaps or improve data resolution), etc. Then GIS (geographic information system), meteorological software or programming tools (such as libraries such as Matplotlib and Plotly of Python) are used to draw the real-time wind map. The generated real-time wind map can clearly show the distribution of wind speed and wind direction, and wind speed and wind direction can be represented by the size and direction of the arrow.
[0064] Based on the real-time wind map of the target detection area, the target indicator data is displayed on the real-time wind map using a suitable annotation method. Among them, it is achieved by adding color-coded points, circles, icons or other graphic elements on the wind map, and the color, size or shape of each element can represent different target indicator values (such as pollutant concentration, number of particles, etc.). The annotation elements are placed on the corresponding detection positions to ensure that the elements are coordinated with the wind speed and wind direction information on the real-time wind map. After completing the data annotation, the analyst can intuitively observe the annotation elements on the real-time wind map to understand the distribution of the target indicator data in the detection area. For example, information such as high-concentration pollutant areas and pollutant diffusion paths can be observed. Optionally, based on the real-time wind map of the annotated data, GIS tools or data analysis software can also be used to further analyze the annotated data to calculate quantitative indicators such as pollutant diffusion speed and impact range. Through the above steps, the real-time wind map of the target detection area can be combined with the target indicator data to intuitively display the distribution of atmospheric pollutants, providing strong support for subsequent pollution source tracking and pollution control.
[0065] Optionally, refer to Figure 3 , after determining the distribution information of the target index data on the real-time wind field map, it also includes:
[0066] S1301, query the historical atmospheric pollutant source tracing record database to determine the historical atmospheric pollutant source tracing records that match the real-time wind field map and distribution information;
[0067] S1302. Extract historical pollution sources from historical atmospheric pollutant source tracing records to serve as reference location points for tracing the source of current target atmospheric pollutants.
[0068] After determining the distribution information of the target indicator data on the real-time wind field map, in order to further narrow the scope of the pollution source and respond quickly, this application queries the historical atmospheric pollutant source traceability record database and extracts relevant information as the traceability reference location point of the current target atmospheric pollutant. Among them, by accessing the database for storing historical atmospheric pollutant source traceability records, the database contains detailed pollution source information, pollutant types, occurrence time, impact range, wind field conditions and other data. When querying the database, according to the characteristics of the real-time wind field map (such as wind speed, wind direction, pollutant distribution pattern) and the distribution information of the target indicator data (such as pollutant concentration peak position, diffusion path, etc.), construct conditions for querying the historical database. These conditions can filter out historical records that are similar or matching the current situation.
[0069] Then submit the constructed query conditions to the database management system to execute the query operation. Filter out the historical atmospheric pollutant source traceability records that best match the current situation from the query results. The screening process can be implemented based on similarity, matching or probability evaluation comparison. Finally, extract key information from the screened historical records, especially the location, type, emission characteristics, etc. of historical pollution sources. This information will serve as the reference location for tracing the current target atmospheric pollutants. Combine the extracted historical pollution source information with the real-time wind field map and target indicator data distribution information to integrate and compare the data to form a more comprehensive basis for pollution source tracing analysis. In this way, the information in the historical atmospheric pollutant source traceability record database can be effectively utilized to provide a strong reference and support for the tracing of the current target atmospheric pollutants.
[0070] S140, constructing a traceability path for the target detection area based on the distribution information, searching for target location points that match the target-related industry information within the set range of the traceability path, and performing pollution source location warning for the target atmospheric pollutants based on the target location points.
[0071] Finally, based on the distribution information of the target indicator data on the real-time wind field map, by constructing a traceability path for the target detection area and further querying the target location points that match the target-related industry information, the pollution source positioning and early warning of the target atmospheric pollutants can be achieved.
[0072] Wherein, optionally constructing a traceability path of the target detection area based on the distribution information includes:
[0073] The distribution path of the target atmospheric pollutants is constructed based on the distribution information, and the road section on the distribution path that matches the wind direction of the real-time wind field map is determined as the tracing path of the target detection area.
[0074] By analyzing the distribution information of the target indicator data on the real-time wind field map, especially the key parameters such as the peak value of pollutant concentration, diffusion direction, and diffusion speed, it is determined whether the distribution direction of the distribution information matches a certain wind direction path on the real-time wind field map based on the wind field data and pollutant distribution information. If so, this matching section is used as the traceability path of the target detection area.
[0075] Reference Figure 4 In the target detection area A, the wind direction a of the corresponding area is obtained according to the real-time wind field map. Similarly, the position b of the distribution information on the wind direction a path is determined according to the distribution information on the real-time wind field map, that is, the distribution information position b matches the wind direction a. Then the section where the distribution information b matches the wind direction a is used as the tracing path of the target detection area. It should be noted that if the number of matches between the distribution information b and the wind direction a is small, the determination of the path can be ignored. The tracing path is determined only when the matching amount between the distribution information b and the wind direction a reaches the set threshold. By drawing possible tracing paths on the real-time wind field map, these paths may cover all directions from which pollutants may come, and take into account changes in wind speed and wind direction as well as the influence of topography.
[0076] Then, within the set range of the traceability path, query the target location points that match the target-related industry information corresponding to the target air pollutant. These location points should be the sources of pollutants such as enterprises or factories. According to the query results, the possible traceability location points are prioritized. The sorting basis may include the distance from the pollution source, the contribution rate of the pollutant concentration, the historical pollution record, etc. The location points with the highest ranking are defined as the target location points, so as to carry out the pollution source location warning of the target air pollutant based on the target location points. Based on the target location point, the analyst can conduct on-site verification to confirm whether there is pollutant emission behavior and whether the emission intensity is consistent with the real-time monitoring data. If it is confirmed that there is a pollution source and its emission behavior is consistent with the monitored pollutant distribution information, the warning information will be issued immediately. The warning information should include the location of the pollution source, the type of pollutant, the concentration range, the possible affected area and the recommended protective measures. After the warning is issued, continue to monitor the target detection area in real time, and pay attention to the changes in the pollutant concentration and the diffusion trend. If the pollutant concentration continues to rise or the diffusion range expands, the emergency plan can be immediately activated and necessary measures can be taken to reduce the emission and diffusion of pollutants. In addition, analytical equipment can share monitoring data and warning information with relevant departments, enterprises and the public in a timely manner so that everyone can jointly address air pollution issues, thereby achieving accurate tracing of air pollutants and early warning prompts.
[0077] Optionally, a traceability path of the target detection area is constructed based on the distribution information, including:
[0078] The pollution concentration gradient change path of the target atmospheric pollutant is constructed based on the distribution information. When the pollution concentration gradient change path matches the wind direction of the real-time wind field map, the pollution concentration gradient change path is determined as the tracing path of the target detection area.
[0079] In the process of constructing the traceability path of the target detection area based on the distribution information, by collecting real-time concentration data of the target atmospheric pollutants, and then using GIS (geographic information system) or other drawing tools, the collected concentration data is drawn into a distribution map to intuitively show the spatial distribution of pollutants. On the distribution map, the gradient changes of pollutant concentration from high to low are identified, and these gradient changes indicate the diffusion direction of the pollutants. By connecting the points of the gradient change of concentration data on the distribution map from high to low, one or more possible distribution paths are formed. Then, the constructed distribution path is compared with the wind direction on the real-time wind field map to find those sections that conform to the law of wind direction diffusion. By screening out the sections that match the wind direction from the comparison results, these sections are used as the traceability path for the diffusion of pollutants. It can be understood that the diffusion characteristics of pollutants may change in gradient along the wind direction. Therefore, by constructing a gradient change distribution path of pollutants, and then by matching the real-time wind direction to construct an accurate traceability path, it can provide strong support for the rapid positioning and effective prevention and control of pollution sources.
[0080] In addition, after locating and warning the pollution source of target air pollutants based on the target location point, it also includes:
[0081] A detection path is constructed based on the wind direction data of the target location point and the real-time wind field map, and the unmanned equipment is driven to collect the pollution concentration change information of the target atmospheric pollutants corresponding to the detection path. The target location point is traced and located based on the pollution concentration change information.
[0082] After locating and warning the pollution source of the target atmospheric pollutants based on the target location point, in order to further confirm and verify the accuracy of the pollution source, the present application constructs a detection path based on the wind direction data of the target location point and the real-time wind field map, and uses unmanned equipment to collect relevant data for source tracing and positioning verification.
[0083] Among them, by extracting the wind direction data around the target location point from the real-time wind field map, special attention is paid to the dominant wind direction and wind speed information. According to the wind direction data and the target location point, one or more detection paths are designed. These paths should extend as far as possible along the direction where pollutants may spread and cover possible pollution source areas. At the same time, considering the influence of factors such as terrain and obstacles on wind direction and wind speed, the paths are reasonably adjusted. According to the needs of the detection task, select suitable unmanned equipment, such as drones, unmanned vehicles or robots. These equipment should have environmental monitoring functions and be able to collect pollution concentration change information of target atmospheric pollutants. Configure corresponding sensors on the unmanned equipment, such as gas detectors, particulate matter detectors, etc., to ensure that the concentration of target atmospheric pollutants can be accurately measured. Set task parameters for the unmanned equipment, including flight / driving routes, sampling frequency, data transmission methods, etc., to ensure that it can collect data according to the predetermined detection path. Then, according to the preset detection path, drive the unmanned equipment to collect data. The unmanned equipment will move along the path and measure the pollution concentration at the preset sampling points. While collecting data, the unmanned equipment transmits the data to the traceability analysis equipment through wireless communication. This allows real-time understanding of changes in pollution concentration at each point on the detection path.
[0084] According to the pollution concentration change information collected by unmanned equipment, the diffusion law and concentration distribution of pollutants on the detection path are analyzed. Then the analysis results are compared with the target location point to verify whether the target location point is indeed the location of the pollution source. If the pollution concentration change trend on the detection path is consistent with the target location point, and there is an obvious concentration peak near the location, the target location point can be considered to be the location of the pollution source. In addition, in addition to the pollution concentration change information, other evidence (such as on-site investigation results, enterprise emission records, etc.) can also be combined for a comprehensive assessment to further improve the accuracy and reliability of source tracing and positioning.
[0085] Furthermore, after driving the unmanned equipment to collect the pollution concentration change information of the target air pollutants corresponding to the detection path, it also includes:
[0086] Based on the pollution concentration change information, it is detected that the pollution concentration difference at the designated location point of the detection path reaches the set threshold, and the pollution source positioning warning of the target atmospheric pollutant is carried out based on the designated location point.
[0087] After driving the unmanned equipment to collect the pollution concentration change information of the target atmospheric pollutants corresponding to the detection path, the pollution concentration change information collected by the unmanned equipment is analyzed in detail, especially focusing on the concentration data of each point on the detection path. On the detection path, the pollution concentration drop between adjacent sampling points or specific sections is calculated. According to historical data, environmental standards or expert opinions, a reasonable pollution concentration drop threshold is set. This threshold is used to determine whether the change in pollution concentration is significant, thereby indicating the presence of a pollution source.
[0088] By screening out sampling points or sections on the detection path where the pollution concentration difference reaches or exceeds the set threshold. From these screened points or sections, further determine one or more designated locations that are most likely to indicate the presence of pollution sources. These locations usually have significant concentration change characteristics, such as sharply increased or decreased concentration values. Once the designated location point is determined and it is confirmed that its pollution concentration difference reaches the set threshold, a pollution source location warning can be issued. The warning information may include the specific location of the designated location point, changes in pollution concentration, possible types of pollution sources, and recommended countermeasures.
[0089] It is understandable that, affected by the real-time wind field, the target air pollutants will diffuse along the wind field direction. Therefore, at the starting point of diffusion, there must be a certain drop in the pollution concentration of the target air pollutants. By detecting this situation, the pollution source of the target air pollutants can be located and warned in a timely and accurate manner, providing strong support for subsequent governance and prevention and control work.
[0090] In the above, by acquiring the atmospheric component data information of the target detection area, the atmospheric component data information is input into the pre-constructed atmospheric pollutant characteristic detection model, and the target atmospheric pollutants and the corresponding target-related industry information of the target detection area are output based on the atmospheric pollutant characteristic detection model. The atmospheric pollutant characteristic detection model pre-constructs training samples based on the atmospheric component characteristic data of different atmospheric pollutants and the corresponding related industry information, and performs model training based on the training samples; determines the target indicator data to be detected based on the target atmospheric pollutants in the target detection area, and collects the target indicator data detected by the collection nodes of each designated detection position in the target detection area; obtains the real-time wind field map of the target detection area, marks the target indicator data on the real-time wind field map based on the designated detection position, and determines the distribution information of the target indicator data on the real-time wind field map; constructs a traceability path for the target detection area based on the distribution information, queries the target location point that matches the target-related industry information within the set range of the traceability path, and performs pollution source location warning for the target atmospheric pollutants based on the target location point. By adopting the above-mentioned technical means, the target atmospheric pollutants can be accurately detected through the atmospheric pollutant characteristic detection model, and then the target atmospheric pollutants can be accurately traced in combination with the real-time wind field. This can improve the efficiency and accuracy of atmospheric pollutant tracing, carry out atmospheric pollution control in a timely manner, and improve the reliability and stability of atmospheric pollution control.
[0091] Embodiment 2:
[0092] Based on the above embodiments, Figure 5 This is a schematic diagram of the structure of an artificial intelligence-based precise source tracing analysis system for atmospheric pollutants provided in Example 2 of this application. Figure 5 The artificial intelligence-based precise source tracing analysis system for atmospheric pollutants provided in this embodiment specifically includes: a model detection module 21, an indicator collection module 22, a distribution analysis module 23 and a source tracing module 24.
[0093] The model detection module 21 is used to obtain atmospheric component data information of the target detection area, input the atmospheric component data information into a pre-constructed atmospheric pollutant characteristic detection model, and output the target atmospheric pollutants and corresponding target associated industry information of the target detection area based on the atmospheric pollutant characteristic detection model. The atmospheric pollutant characteristic detection model constructs training samples based on atmospheric component characteristic data of different atmospheric pollutants and corresponding associated industry information in advance, and performs model training based on the training samples.
[0094] The index collection module 22 is used to determine the target index data to be detected based on the target air pollutants in the target detection area, and collect the target index data detected by the collection nodes at each designated detection position in the target detection area;
[0095] The distribution analysis module 23 is used to obtain a real-time wind field map of the target detection area, mark the target index data on the real-time wind field map based on the designated detection position, and determine the distribution information of the target index data on the real-time wind field map;
[0096] The tracing module 24 constructs a tracing path for the target detection area based on the distribution information, queries for a target location point that matches the target-related industry information within a set range of the tracing path, and performs pollution source location warning for the target atmospheric pollutants based on the target location point.
[0097] Specifically, after collecting the target index data detected by the collection nodes at each designated detection position in the target detection area, it also includes:
[0098] Based on the target indicator data and the target air pollutants, a reinforcement learning sample is constructed, and the air pollutant feature detection model is trained based on the reinforcement learning sample. Specifically,
[0099] Specifically, after determining the distribution information of the target index data on the real-time wind field map, it also includes:
[0100] Query the historical atmospheric pollutant source tracing record database, determine the historical atmospheric pollutant source tracing records that match the real-time wind field map and distribution information, and extract the historical pollution sources of the historical atmospheric pollutant source tracing records as the tracing reference location points for the current target atmospheric pollutants.
[0101] Specifically, the traceability path of the target detection area is constructed based on the distribution information, including:
[0102] The distribution path of the target atmospheric pollutants is constructed based on the distribution information, and the road section on the distribution path that matches the wind direction of the real-time wind field map is determined as the tracing path of the target detection area.
[0103] Specifically, the traceability path of the target detection area is constructed based on the distribution information, including:
[0104] The pollution concentration gradient change path of the target atmospheric pollutant is constructed based on the distribution information. When the pollution concentration gradient change path matches the wind direction of the real-time wind field map, the pollution concentration gradient change path is determined as the tracing path of the target detection area.
[0105] Specifically, after locating and warning the pollution source of the target air pollutant based on the target location point, it also includes:
[0106] A detection path is constructed based on the wind direction data of the target location point and the real-time wind field map, and the unmanned equipment is driven to collect the pollution concentration change information of the target atmospheric pollutants corresponding to the detection path. The target location point is traced and located based on the pollution concentration change information.
[0107] Specifically, after driving the unmanned equipment to collect the pollution concentration change information of the target air pollutants corresponding to the detection path, it also includes:
[0108] Based on the pollution concentration change information, it is detected that the pollution concentration difference at the designated location point of the detection path reaches the set threshold, and the pollution source positioning warning of the target atmospheric pollutant is carried out based on the designated location point.
[0109] In the above, by acquiring the atmospheric component data information of the target detection area, the atmospheric component data information is input into the pre-constructed atmospheric pollutant characteristic detection model, and the target atmospheric pollutants and the corresponding target-related industry information of the target detection area are output based on the atmospheric pollutant characteristic detection model. The atmospheric pollutant characteristic detection model pre-constructs training samples based on the atmospheric component characteristic data of different atmospheric pollutants and the corresponding related industry information, and performs model training based on the training samples; determines the target indicator data to be detected based on the target atmospheric pollutants in the target detection area, and collects the target indicator data detected by the collection nodes of each designated detection position in the target detection area; obtains the real-time wind field map of the target detection area, marks the target indicator data on the real-time wind field map based on the designated detection position, and determines the distribution information of the target indicator data on the real-time wind field map; constructs a traceability path for the target detection area based on the distribution information, queries the target location point that matches the target-related industry information within the set range of the traceability path, and performs pollution source location warning for the target atmospheric pollutants based on the target location point. By adopting the above-mentioned technical means, the target atmospheric pollutants can be accurately detected through the atmospheric pollutant characteristic detection model, and then the target atmospheric pollutants can be accurately traced in combination with the real-time wind field. This can improve the efficiency and accuracy of atmospheric pollutant tracing, carry out atmospheric pollution control in a timely manner, and improve the reliability and stability of atmospheric pollution control.
[0110] The artificial intelligence-based precise source tracing and analysis system for atmospheric pollutants provided in Example 2 of the present application can be used to execute the artificial intelligence-based precise source tracing and analysis method for atmospheric pollutants provided in Example 1 above, and has corresponding functions and beneficial effects.
[0111] Embodiment three:
[0112] Embodiment 3 of the present application provides an electronic device, referring to Figure 6The electronic device includes: a processor 31, a memory 32, a communication module 33, an input device 34 and an output device 35. The number of processors in the electronic device can be one or more, and the number of memories in the electronic device can be one or more. The processor, memory, communication module, input device and output device of the electronic device can be connected via a bus or other means.
[0113] As a computer-readable storage medium, the memory can be used to store software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the precise traceability analysis method of atmospheric pollutants based on artificial intelligence described in any embodiment of the present application (for example, the model detection module, index acquisition module, distribution analysis module and traceability module in the precise traceability analysis system of atmospheric pollutants based on artificial intelligence). The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function; the storage data area may store data created according to the use of the device, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network and a combination thereof.
[0114] The communication module is used for data transmission.
[0115] The processor executes various functional applications and data processing of the device by running software programs, instructions and modules stored in the memory, thereby realizing the above-mentioned artificial intelligence-based precise source tracing analysis method of atmospheric pollutants.
[0116] The input device can be used to receive input digital or character information and generate key signal input related to user settings and function control of the device. The output device can include display devices such as display screens.
[0117] The electronic device provided above can be used to execute the artificial intelligence-based precise source tracing analysis method for atmospheric pollutants provided in the above-mentioned embodiment 1, and has corresponding functions and beneficial effects.
[0118] Embodiment 4:
[0119] The embodiment of the present application also provides a storage medium containing computer executable instructions, which are used to execute an artificial intelligence-based precise source tracing analysis method for atmospheric pollutants when executed by a computer processor. The artificial intelligence-based precise source tracing analysis method for atmospheric pollutants includes: obtaining atmospheric component data information of a target detection area, inputting the atmospheric component data information into a pre-constructed atmospheric pollutant characteristic detection model, and outputting target atmospheric pollutants and corresponding target associated industry information of the target detection area based on the atmospheric pollutant characteristic detection model, wherein the atmospheric pollutant characteristic detection model constructs training samples based on atmospheric component characteristic data of different atmospheric pollutants and corresponding associated industry information in advance, and .... Model training is performed on the training samples; target indicator data to be detected is determined based on the target atmospheric pollutants in the target detection area, and the target indicator data detected by the collection nodes at each designated detection position in the target detection area is collected; a real-time wind field map of the target detection area is obtained, and the target indicator data is marked on the real-time wind field map based on the designated detection position, and distribution information of the target indicator data on the real-time wind field map is determined; a traceability path of the target detection area is constructed based on the distribution information, and a target location point matching the target-related industry information is queried within a set range of the traceability path, and a pollution source location warning of the target atmospheric pollutants is performed based on the target location point.
[0120] Storage medium - any of various types of memory devices or storage devices. The term "storage medium" is intended to include: installation media, such as CD-ROM, floppy disk or tape device; computer system memory or random access memory, such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory, such as flash memory, magnetic media (such as hard disk or optical storage); registers or other similar types of memory elements, etc. Storage media may also include other types of memory or combinations thereof. In addition, the storage medium may be located in the first computer system in which the program is executed, or may be located in a different second computer system, which is connected to the first computer system via a network (such as the Internet). The second computer system can provide program instructions to the first computer for execution. The term "storage medium" may include two or more storage media residing in different locations (for example, in different computer systems connected by a network). The storage medium may store program instructions (for example, embodied as a computer program) that can be executed by one or more processors.
[0121] Of course, the storage medium containing computer executable instructions provided in the embodiments of the present application is not limited to the above-mentioned precise source tracing and analysis method of atmospheric pollutants based on artificial intelligence, but can also execute related operations in the precise source tracing and analysis method of atmospheric pollutants based on artificial intelligence provided in any embodiment of the present application.
[0122] The artificial intelligence-based precise source tracing and analysis system for atmospheric pollutants, storage medium and electronic device provided in the above embodiments can execute the artificial intelligence-based precise source tracing and analysis method for atmospheric pollutants provided in any embodiment of the present application. For technical details not described in detail in the above embodiments, please refer to the artificial intelligence-based precise source tracing and analysis method for atmospheric pollutants provided in any embodiment of the present application.
[0123] The above are only preferred embodiments of the present application and the technical principles used. The present application is not limited to the specific embodiments described herein, and various obvious changes, readjustments and substitutions that can be made by those skilled in the art will not deviate from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the claims.
Claims
1. An artificial intelligence-based precise source tracing analysis method for atmospheric pollutants, characterized in that: include: Acquire atmospheric component data information of a target detection area, input the atmospheric component data information into a pre-constructed atmospheric pollutant characteristic detection model, output target atmospheric pollutants and corresponding target associated industry information of the target detection area based on the atmospheric pollutant characteristic detection model, the atmospheric pollutant characteristic detection model constructs training samples based on atmospheric component characteristic data of different atmospheric pollutants and corresponding associated industry information in advance, and performs model training based on the training samples; Determine the target index data to be detected based on the target air pollutants in the target detection area, and collect the target index data detected by the collection nodes at each designated detection position in the target detection area; Acquire a real-time wind field map of the target detection area, mark the target index data on the real-time wind field map based on the designated detection position, and determine distribution information of the target index data on the real-time wind field map; A traceability path for the target detection area is constructed based on the distribution information, a target location point matching the target-associated industry information is searched within a set range of the traceability path, and a pollution source location warning for the target atmospheric pollutant is performed based on the target location point.
2. The method for accurate source tracing of atmospheric pollutants based on artificial intelligence according to claim 1 is characterized in that: The constructing the tracing path of the target detection area based on the distribution information includes: The distribution path of the target atmospheric pollutant is constructed according to the distribution information, and the section on the distribution path that matches the wind direction of the real-time wind field map is determined as the tracing path of the target detection area.
3. The method for accurate source tracing of atmospheric pollutants based on artificial intelligence according to claim 1 is characterized in that: The constructing the tracing path of the target detection area based on the distribution information includes: A pollution concentration gradient change path of the target atmospheric pollutant is constructed based on the distribution information. When the pollution concentration gradient change path matches the wind direction of the real-time wind field map, the pollution concentration gradient change path is determined as the tracing path of the target detection area.
4. The method for accurate source tracing of atmospheric pollutants based on artificial intelligence according to any one of claims 1 to 3, characterized in that: After the pollution source positioning and early warning of the target air pollutant is performed based on the target location point, the method further includes: A detection path is constructed according to the wind direction data of the target location point and the real-time wind field map, and unmanned equipment is driven to collect pollution concentration change information corresponding to the target atmospheric pollutants in the detection path, and the target location point is traced and located based on the pollution concentration change information.
5. The method for accurate source tracing of atmospheric pollutants based on artificial intelligence according to claim 4 is characterized in that: After driving the unmanned equipment to collect the pollution concentration change information corresponding to the target air pollutant in the detection path, the method further includes: Based on the pollution concentration change information, it is detected that the pollution concentration difference at the designated position point of the detection path reaches a set threshold, and a pollution source positioning warning of the target atmospheric pollutant is performed based on the designated position point.
6. The method for accurate source tracing of atmospheric pollutants based on artificial intelligence according to claim 1 is characterized in that: After determining the distribution information of the target index data on the real-time wind field map, the method further includes: Query the historical atmospheric pollutant source tracing record database, determine the historical atmospheric pollutant source tracing records that match the real-time wind field map and the distribution information, and extract the historical pollution sources of the historical atmospheric pollutant source tracing records to serve as the tracing reference location points for the current target atmospheric pollutants.
7. The method for accurate source tracing of atmospheric pollutants based on artificial intelligence according to claim 1 is characterized in that: After collecting the target indicator data detected by the collection nodes at each designated detection position in the target detection area, the method further includes: A reinforcement learning sample is constructed based on the target indicator data and the target air pollutant, and the air pollutant feature detection model is trained based on the reinforcement learning sample.
8. An artificial intelligence-based precise source tracing analysis system for atmospheric pollutants, characterized in that: include: A model detection module is used to obtain atmospheric component data information of a target detection area, input the atmospheric component data information into a pre-constructed atmospheric pollutant characteristic detection model, and output target atmospheric pollutants and corresponding target associated industry information of the target detection area based on the atmospheric pollutant characteristic detection model. The atmospheric pollutant characteristic detection model constructs training samples based on atmospheric component characteristic data of different atmospheric pollutants and corresponding associated industry information in advance, and performs model training based on the training samples; An index collection module, used to determine target index data to be detected based on the target atmospheric pollutants in the target detection area, and collect the target index data detected by the collection nodes at each designated detection position in the target detection area; A distribution analysis module, used to obtain a real-time wind field map of the target detection area, mark the target index data on the real-time wind field map based on the designated detection position, and determine the distribution information of the target index data on the real-time wind field map; The tracing module constructs a tracing path for the target detection area based on the distribution information, queries for a target location point that matches the target-related industry information within a set range of the tracing path, and performs pollution source location warning for the target atmospheric pollutants based on the target location point.
9. An electronic device, characterized in that: include: memory and one or more processors; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the artificial intelligence-based precise source tracing analysis method for atmospheric pollutants as described in any one of claims 1-7.
10. A storage medium containing computer executable instructions, characterized in that: The computer executable instructions, when executed by a computer processor, are used to execute the artificial intelligence-based precise source tracing analysis method for atmospheric pollutants as described in any one of claims 1-7.
Citation Information
Patent Citations
Air pollution monitoring system based on Beidou positioning
CN104849415A
Monitoring method and related equipment for air quality
CN107607450A
Method and device for tracing air pollutants, equipment and storage medium
CN110687255A
Industrial park atmospheric pollutant diffusion simulating and tracing method
CN111537023A
Pollutant monitoring method and system based on geographic information system
CN111781306A
Cited By
Method and system for detecting excessive emission of atmospheric pollutants
CN121068862A
Method and system for detecting over-standard emission of atmospheric pollutants
CN121068862B
Port air quality monitoring method and system based on multi-source data analysis
CN122046222A
Port air quality monitoring method and system based on multi-source data analysis
CN122046222B
