A system for identifying industrial aggregated composite pollutants based on big data

Abstract

The present invention discloses a system for identifying complex pollutants in industrial agglomerations based on big data, and relates to the technical field of industrial pollutants. The system comprises an acquisition module, a processing module and an output module, wherein the acquisition module stores a basic information data set of an industrial park, a chemical knowledge information data set, a factory emission information data set and an emission index information data set. The efficient collaboration of the acquisition module, the processing module and the output module of the present invention realizes the comprehensive monitoring and accurate identification of complex pollutants in industrial agglomeration areas. The acquisition module integrates multi-source data and provides a rich basis for analysis. The predictive analysis model in the processing module uses a long short-term memory network to extract key features, and through continuous optimization and updating, ensures the accuracy and timeliness of the model, can accurately identify the conditions for the generation of complex pollutants, predict their development trends, and provide timely feedback on factory information with excessive emissions.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial pollutants, and in particular to an industrial aggregated composite pollutant identification system based on big data. Background Art

[0002] Traditional pollutant identification methods mainly rely on manual sampling and laboratory analysis. This method has many limitations, such as limited sampling points, untimely data acquisition, and complex analysis process. Therefore, accurately identifying complex pollutants in industrial clusters has become an important task in environmental management and pollution control.

[0003] The publication number is CN114970722A, and the name is a pollutant identification method, device, electronic device and storage medium. The method obtains ultraviolet-visible spectrum data of the solution to be tested and converts the ultraviolet-visible spectrum data into a spectrum vector; the spectrum vector is input into a preset noise reduction network model for noise reduction processing to obtain a noise-reduced spectrum vector; the spectrum vector and the noise-reduced spectrum vector are combined to obtain a feature vector, and the feature vector is input into a preset classification network model for pollutant identification to obtain the category of excessive pollutants contained in the solution to be tested; the preset noise reduction network model and the preset classification network model can be used to improve the robustness, efficiency and accuracy of pollutant identification, reduce the requirements for water quality testing equipment, and thus can identify pollutants at low cost and high efficiency.

[0004] The above technology focuses on improving the robustness, efficiency and accuracy of pollutant identification through preset noise reduction network models and preset classification network models, and reduces the requirements for water quality testing equipment, thereby achieving low-cost and efficient pollutant identification. However, in the identification scenario of complex pollutants in industrial agglomerations, this method still has some shortcomings. First, this method is mainly suitable for water quality testing, and the identification of complex pollutants in air quality may have limited effect, because the composition of pollutants in the air is more complex, and their concentration and diffusion mode are greatly affected by meteorological conditions. Secondly, this method relies on laboratory analysis. For real-time, large-scale air quality monitoring of industrial agglomerations, the timeliness of data acquisition and processing may not meet the requirements. Finally, this method mainly focuses on the identification of pollutants exceeding the standard, and lacks in-depth analysis of the formation mechanism and influencing factors of complex pollutants, making it difficult to provide targeted pollution control strategies. Summary of the invention

[0005] The purpose of the present invention is to provide an industrial aggregate complex pollutant identification system based on big data to solve the problems raised in the above background technology.

[0006] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an industrial aggregated composite pollutant identification system based on big data, comprising a collection module, a processing module and an output module;

[0007] The acquisition module stores the industrial park basic information data set, the chemical knowledge information data set, the factory emission information data set and the emission index information data set. The acquisition module sets monitoring equipment at key locations in the industrial park to regularly monitor the air quality of the industrial park to obtain a monitoring information data set and a monitoring equipment information data set. The acquisition module sets a meteorological monitoring station in the industrial park to monitor the weather in the industrial park to obtain a meteorological information data set. The industrial park basic information data set includes an air basic information data set, and the monitoring information data set includes a real-time monitoring information data set and a historical monitoring information data set.

[0008] A prediction and analysis model is provided in the processing module. The processing module derives a composite pollutant information data set based on the factory emission information data set and the meteorological information data set in the acquisition module according to the pollutant derivation method, and extracts the generation conditions of the composite pollutants based on the composite pollutant information data set to obtain a generation condition information set. The processing module adjusts the monitoring frequency of the monitoring equipment based on the generation condition information set and the meteorological information data set in combination with the key monitoring method. The processing module matches and associates the factory emission information data set with the emission index information data set to obtain a matching information set. The prediction model obtains an air quality report based on the real-time monitoring information data set and the matching information set, and obtains a predicted composite pollutant report based on the air quality report and the meteorological information data set. Finally, the processing module matches the air quality report with the matching information set to obtain an information set of factories with excessive emissions.

[0009] The output module outputs identification reports based on air quality reports, factories with excessive emissions, and forecasted composite pollutant reports;

[0010] The predictive analysis model includes:

[0011] S1: The industrial park basic information dataset, chemical knowledge information dataset, factory emission information dataset, emission index information dataset, meteorological information dataset and monitoring information dataset are preprocessed to obtain standard data, which are divided into training set, validation set and test set, and a database is set up to integrate and store the standard data;

[0012] S2: Select the long short-term memory network as the feature extractor to extract the features of the training set and extract the target features in the training set;

[0013] S3: Train the feature extractor based on the training set to obtain an initial model for analyzing the air quality of the industrial park and predicting the generation of composite pollutants;

[0014] S4: Train and optimize the initial model to obtain a predictive analysis model, and regularly update the predictive analysis model.

[0015] Furthermore, the training and optimization method includes:

[0016] A1: Model evaluation: evaluate the trained initial model based on the validation set to obtain the evaluation results, which are used to analyze the performance of the model and identify existing problems and improvement points.

[0017] A2: Model optimization: optimize the model based on the evaluation results and use the grid search method to tune the parameters;

[0018] A3: Model verification: verify the optimized model based on the test set to ensure the generalization ability of the model on unknown data;

[0019] A4: Model update: Update the matching model regularly and use incremental learning methods to update the model in real time.

[0020] Furthermore, the pollutant derivation method includes:

[0021] B1: Determine the type and chemical composition of a single pollutant based on the factory emission information data set to obtain the primary pollutant information set;

[0022] B2: Determine wind speed, wind direction, temperature, humidity and atmospheric pressure based on the meteorological information data set, thereby obtaining a meteorological parameter information set;

[0023] B3: The chemical reactivity information set is obtained by analyzing each pollutant gas in the primary pollutant information set, and then combined with the meteorological parameter information set and the chemical reaction equations in the chemical knowledge information data set for comparison and analysis to obtain the secondary pollutant information set;

[0024] B4: A composite pollutant information dataset is obtained by integrating the primary pollutant information set, meteorological parameter information set and secondary pollutant information set.

[0025] Furthermore, the analysis method of the chemical reactivity information set includes:

[0026] C1: Based on the chemical composition of each pollutant gas in the primary pollutant information set and the chemical composition of the air in the air basic information data set, the chemical composition information set is integrated to obtain the structure and electronic configuration of each chemical component, thereby obtaining the potential reaction information set;

[0027] C2: Based on the comparison between the chemical composition information set and the chemical knowledge information data set, the chemical properties and reaction conditions of each chemical component in the chemical composition information set are obtained, thereby obtaining the chemical reaction type information set;

[0028] C3: Compare the meteorological parameter information set with the reaction conditions of each chemical component to obtain the meteorological parameters that affect the reaction rate and occurrence probability, thereby obtaining the target meteorological parameter set;

[0029] C4: Integrate C1-C3 to obtain the chemical reactivity information set.

[0030] Furthermore, the method for extracting the generation condition information set includes:

[0031] D1: Based on each composite pollutant in the composite pollutant information data set, identify the key conditions required for the generation of each composite pollutant and obtain the key condition information set;

[0032] D2: Based on the key condition information set, the integrated information set is obtained by classifying and sorting according to meteorological parameters, pollutant types, concentrations and reaction times;

[0033] D3: After verification based on the integrated information set, the generation condition information set is finally obtained.

[0034] Furthermore, the method for determining the key position is:

[0035] T1: Based on the industrial park basic information data set and matching information set, GIS technology is used to determine the location of key emission sources, and based on the meteorological information data set, meteorological software is used to determine the diffusion path and range of pollutants;

[0036] T2: Based on the basic information data set of the industrial park, the diffusion path and range of pollutants, GIS technology is used to determine the pollutant accumulation area; based on the historical monitoring information data set, GIS technology is used to determine the pollutant diffusion law and historical monitoring location;

[0037] T3: Based on the locations of key emission sources, pollutant accumulation areas and historical monitoring locations, GIS technology is used to obtain key location information sets to obtain key locations.

[0038] Furthermore, the key monitoring method includes:

[0039] Y1: Based on the meteorological information data set, the current meteorological parameters are obtained, and the target meteorological parameters that are most similar to the current meteorological parameters are matched with the target meteorological parameter set, so as to obtain the corresponding target meteorological parameter set;

[0040] Y2: Acquire the corresponding generation condition information set based on matching the target meteorological parameter set with the generation condition information set;

[0041] Y3: Based on the matching of the corresponding generation condition information set and the composite pollutant information data set, the corresponding composite pollutant information set is obtained, and then the characteristics of each composite pollutant in the corresponding composite pollutant information set are evaluated to obtain an evaluation information set, and each evaluation information in the evaluation information set is prioritized;

[0042] Y4: Based on the sorted evaluation information set, the monitoring frequency of the monitoring equipment is adjusted through the remote adjustment software to increase the monitoring frequency for high-priority composite pollutants and reduce the monitoring frequency for low-priority composite pollutants.

[0043] Furthermore, the processing module is connected to a control terminal, and the monitoring equipment, meteorological monitoring station, acquisition module, processing module and output module at key positions are all connected through a proprietary local area network. The method for creating the proprietary local area network includes:

[0044] N1: Information acquisition, obtaining basic networking information of the acquisition module, processing module, output module, monitoring equipment and meteorological monitoring station, the basic networking information including the equipment IP;

[0045] N2: Device connection, the control terminal automatically allocates the communication channel to match the IP of the acquisition module, the IP of the processing module, the IP of the output module, the IP of the monitoring device and the IP of the meteorological monitoring station;

[0046] N3: Configure the network, control the terminal to configure the corresponding network services and security measures for the communication channel, and create a private LAN.

[0047] Furthermore, the monitoring equipment includes an air quality monitoring station, and the air quality monitoring station is equipped with multiple gas analyzers, which can monitor multiple pollutants simultaneously.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] A big data-based industrial cluster composite pollutant identification system realizes comprehensive monitoring and accurate identification of composite pollutants in industrial clusters through efficient collaboration of acquisition modules, processing modules and output modules. The acquisition module integrates multi-source data and provides a rich analysis basis. The predictive analysis model in the processing module uses long short-term memory networks to extract key features and ensures the accuracy and timeliness of the model through continuous optimization and updating. It can accurately identify the generation conditions of composite pollutants, predict their development trends, and promptly feedback information on factories with excessive emissions. Compared with traditional laboratory analysis and preset model methods, the system is more comprehensive, efficient and intelligent, and can better cope with the challenges of identifying and managing composite pollutants in industrial clusters, providing strong support for environmental management and pollution control, which not only improves the efficiency of environmental supervision, but also effectively promotes the green and sustainable development of industrial clusters.

[0050] At the same time, the system can accurately identify and predict the types and production conditions of complex pollutants through the pollutant derivation method, combined with factory emission information, meteorological parameters and chemical knowledge information. Through the analysis of chemical reactivity information sets and the extraction of production condition information sets, the system deeply analyzes the formation mechanism and influencing factors of complex pollutants, providing strong support for pollution control. In addition, the system uses GIS technology to determine key locations, and realizes accurate monitoring of high-priority complex pollutants through key monitoring methods, which improves the efficiency and pertinence of monitoring. Compared with existing technologies, the system is not only suitable for air quality monitoring, but also can process large amounts of data in real time, providing a more comprehensive, efficient and intelligent solution for environmental management of industrial clusters. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the principle structure of the present invention;

[0052] Figure 2 It is a schematic diagram of the principle structure of the prediction analysis model of the present invention;

[0053] Figure 3 It is a schematic diagram of the principle structure of the pollutant derivation method of the present invention;

[0054] Figure 4 It is a schematic diagram of the industrial park of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0056] Accurately identifying and controlling complex pollutants in industrial park management is the key to ensuring environmental safety and sustainable industrial production. The industrial aggregate complex pollutant identification system based on big data provided by the present invention is suitable for industrial parks that need to continuously monitor air quality and predict the risks of complex pollutants. When the system is working, it is necessary to ensure that the network of the industrial park is unobstructed so as to transmit monitoring data in real time. At the same time, the meteorological monitoring station should be set at a location that can reflect the overall meteorological conditions of the park. A stable power supply is also required to maintain the normal operation of the monitoring equipment and processing modules. Operators need to have basic computer operation and data analysis capabilities, and receive relevant safety training to ensure that response measures can be taken quickly in emergency situations. At the same time, the system should have a high degree of security to protect sensitive data and avoid unauthorized access.

[0057] like Figure 1-Figure 4 As shown, the present invention provides a technical solution: an industrial aggregated composite pollutant identification system based on big data, comprising a collection module, a processing module and an output module;

[0058] The acquisition module stores the industrial park basic information data set, chemical knowledge information data set, factory emission information data set and emission index information data set. The acquisition module sets monitoring equipment at key locations in the industrial park to regularly monitor the air quality of the industrial park to obtain monitoring information data sets and monitoring equipment information data sets. The acquisition module sets up a meteorological monitoring station in the industrial park to monitor the weather in the industrial park to obtain a meteorological information data set. The industrial park basic information data set includes an air basic information data set, and the monitoring information data set includes a real-time monitoring information data set and a historical monitoring information data set.

[0059] It should be noted that the industrial park basic information data set: factory layout / production process / equipment type / park map, air basic information data set: air quality benchmark value / common pollutant types / pollutant concentration range, chemical knowledge information data set: chemical substance properties / chemical reaction equations / reaction conditions / chemical reaction rate constants, factory emission information data set: emission port location / emission type / emission concentration / emission time, emission index information data set: emission standards / violation records / emission permit information, real-time monitoring information data set: pollutant concentration / monitoring time, historical monitoring information data set: past pollutant concentration data and monitoring equipment operation status records, meteorological information data set: real-time meteorological data / meteorological forecast data / meteorological statistical analysis data, regular monitoring is once an hour, and each monitoring time is 5-10 minutes.

[0060] A prediction and analysis model is provided in the processing module. The processing module derives a composite pollutant information data set based on the factory emission information data set and the meteorological information data set in the acquisition module according to the pollutant derivation method, and extracts the generation conditions of the composite pollutants based on the composite pollutant information data set to obtain a generation condition information set. The processing module adjusts the monitoring frequency of the monitoring equipment based on the generation condition information set and the meteorological information data set in combination with the key monitoring method. The processing module matches and associates the factory emission information data set with the emission index information data set to obtain a matching information set. The prediction model obtains an air quality report based on the real-time monitoring information data set and the matching information set, and obtains a predicted composite pollutant report based on the air quality report and the meteorological information data set. Finally, the processing module matches the air quality report with the matching information set to obtain an information set of factories with excessive emissions.

[0061] It should be noted that in the derivation of the composite pollutant information data set, the processing module first reads the factory emission information data set and the meteorological information data set, and uses the pollutant derivation method, combined with the chemical reaction equations and reaction conditions in the chemical knowledge information data set, to analyze the chemical reactions that may occur in the chemical substances emitted by the factory under specific meteorological conditions, thereby deducing the types and concentrations of composite pollutants that may be produced, and storing the derivation results in the composite pollutant information data set. The processing module analyzes the production conditions of various composite pollutants from the composite pollutant information data set, such as specific temperature ranges, humidity conditions, wind speed and direction, etc., and organizes these conditions into a production condition information set for subsequent use in frequency adjustment of monitoring equipment. The processing module reads the production condition information set and the meteorological information data set, and combines the key monitoring method to analyze the areas and time periods that need to be focused on according to the changes in the production conditions and meteorological conditions, and then determines the production conditions and the concentrations of the composite pollutants. Information set, adjust the monitoring frequency of monitoring equipment in these areas and time periods to ensure that changes in pollutants can be captured more accurately. The processing module matches and associates the factory emission information data set with the emission indicator information data set, checks whether the emissions of each factory exceed the prescribed emission standards, and records the factory information that exceeds the standard, such as the factory name, type of pollutants exceeding the standard, and time of exceeding the standard, to form a matching information set. The prediction model reads the real-time monitoring information data set and the matching information set, analyzes the current air quality status, including pollutant concentration, meteorological parameters, etc., generates an air quality report, and combines the meteorological information data set to predict the air quality trend and possible composite pollutant risks in the future, and generates a predicted composite pollutant report. The processing module matches the air quality report with the matching information set, identifies the factories with excessive emissions, and records relevant information, such as the factory name, type of pollutants exceeding the standard, and time of exceeding the standard.

[0062] The output module outputs identification reports based on air quality reports, factories with excessive emissions, and forecasted composite pollutant reports;

[0063] like Figure 2 As shown, the prediction and analysis model includes: S1: pre-processing the industrial park basic information data set, chemical knowledge information data set, factory emission information data set, emission index information data set, meteorological information data set and monitoring information data set to obtain standard data, dividing the standard data into training set, verification set and test set, and setting a database to integrate and store the standard data, S2: selecting a long short-term memory network as a feature extractor to extract features from the training set, and extracting target features from the training set, S3: training the feature extractor based on the training set to obtain an initial model for analyzing the air quality of the industrial park and predicting the generation of complex pollutants, S4: training and optimizing based on the initial model to obtain a prediction and analysis model, and regularly updating the prediction and analysis model.

[0064] It should be noted that the acquisition module collects data from various data sources (such as industrial park basic information data set, factory emission information data set, emission index information data set, meteorological information data set and monitoring information data set), integrates these data into a unified format, removes duplicate or inconsistent data items, cleans missing values, outliers, erroneous data, etc. in the data, and may handle these problems by interpolation, deletion or correction. According to the type and range of the data, standardization or normalization is performed to ensure that all features are numerically comparable. The cleaned and standardized data is divided into training set, validation set and test set. Usually, the training set is used for model training, and the validation set is used for hyperparameter adjustment and to prevent overfitting. The test set is used to evaluate the performance of the model. A database or data warehouse is set up to store standard data to ensure the security and accessibility of the data. Considering the characteristics of time series data and long-term dependencies, LSTM is selected as the feature extractor. According to the characteristics of the data and prediction requirements, the architecture of the LSTM network is designed, including the input layer, hidden layer (including LSTM units) and output layer. The training set data is input into the LSTM network. Through the training process of the network, the target features in the data are automatically extracted. The mean square error is selected as the loss function of the prediction task. The gradient descent optimization algorithm is selected to minimize the loss function. The training set data is used to train the LSTM network. The gradient is calculated and the network parameters are updated through the back propagation algorithm.

[0065] The training and optimization methods include: A1: Model evaluation, evaluating the trained initial model based on the validation set to obtain the evaluation results, which are used to analyze the performance of the model and find out the existing problems and improvement points; A2: Model optimization, optimizing the model based on the evaluation results, using the grid search method for parameter tuning; A3: Model verification, verifying the optimized model based on the test set to ensure the generalization ability of the model on unknown data; A4: Model update, regularly updating the matching model, and using the incremental learning method to update the model in real time.

[0066] It should be noted that appropriate evaluation indicators should be selected according to the prediction task, such as accuracy, recall, F1 score, mean square error (MSE), etc. The trained initial model should be evaluated using the validation set data, the validation set data should be input into the model to obtain the prediction results, the validation set data should be input into the model to obtain the prediction results, the evaluation results should be analyzed, the performance of the model should be evaluated, and problems with the model should be found, such as overfitting, underfitting, unreasonable parameter settings, etc. Improvements should be found based on the problems to provide guidance for the next step of model optimization. Based on the results of the model evaluation, the optimization goals should be determined, such as improving accuracy and reducing errors. The hyperparameters of the model should be tuned using the grid search method, the range and step size of the hyperparameters should be set, a grid of hyperparameters should be generated, each hyperparameter combination should be trained, and the performance of the model on the validation set should be evaluated. The hyperparameter combination with the best performance should be selected as the parameters of the final model. In addition to grid search, other hyperparameter tuning methods such as random search and Bayesian optimization can also be tried. Depending on the characteristics of the problem, you can consider increasing Add structural adjustments such as the depth, width or complexity of the model, prepare test set data, ensure that the test set is consistent with the training set and validation set in data distribution, but completely independent, input the test set data into the optimized model, obtain the prediction results, compare the prediction results with the true labels of the test set, calculate the evaluation indicators, and evaluate the generalization ability of the model on unknown data. If the performance of the model on the test set is similar to or better than that on the validation set, it means that the model has good generalization ability. Regularly collect new data, including basic information of the industrial park, factory emission information, meteorological information, etc. Use incremental learning methods to update the model in real time, add new data to the training set, and fine-tune the model based on the new training set. During the fine-tuning process, a smaller learning rate or regularization parameter can be used to prevent the model from overfitting to the new data. After the model is updated, evaluate and verify the model again to ensure that the performance of the model has not declined. If the model performance declines, consider using more data or adjusting the optimization strategy to further improve the model.

[0067] like Figure 3 As shown, the pollutant derivation method includes: B1: determining the type and chemical composition of a single pollutant based on the factory emission information data set, thereby obtaining a primary pollutant information set; B2: determining wind speed, wind direction, temperature, humidity and atmospheric pressure based on the meteorological information data set, thereby obtaining a meteorological parameter information set; B3: obtaining a chemical reactivity information set based on the analysis of each pollutant gas in the primary pollutant information set, and then combining the meteorological parameter information set and using the chemical reaction equations in the chemical knowledge information data set for comparison and analysis to obtain a secondary pollutant information set; B4: obtaining a composite pollutant information data set based on the integration of the primary pollutant information set, the meteorological parameter information set and the secondary pollutant information set.

[0068] It should be noted that the original data, including information such as pollutant types, emission concentrations, and chemical compositions, are extracted from the factory emission information dataset. The data is cleaned to remove duplicate, erroneous, or irrelevant information. Pollutants are classified according to their types and chemical compositions, and each pollutant is given a unique identifier to construct a primary pollutant information set, which contains detailed information such as the identifier, type, chemical composition, and emission concentration of each pollutant. Real-time or historical meteorological data, including wind speed, wind direction, temperature, humidity, and atmospheric pressure, are obtained from the meteorological information dataset. As needed, the meteorological data is converted into a format suitable for analysis (such as time series data). The data is standardized so that it can be uniformly compared in subsequent analysis. The converted and standardized meteorological data are integrated to construct a meteorological parameter information set, which contains meteorological parameter information such as wind speed, wind direction, temperature, humidity, and atmospheric pressure at each time point. Chemical reactivity analysis is performed on each pollutant gas in the primary pollutant information set to determine its chemical reactivity and possible The reaction products of the primary pollutants are combined with the meteorological parameter information set, and the chemical reaction equations in the chemical knowledge information data set are used for comparative analysis to determine the possible chemical reactions of the primary pollutants and the types of secondary pollutants generated under specific meteorological conditions. The types, production conditions, concentrations and other information of the secondary pollutants obtained by analysis are integrated to construct a secondary pollutant information set, which contains detailed information on each secondary pollutant and its relationship with the primary pollutants and meteorological parameters. The primary pollutant information set, the meteorological parameter information set and the secondary pollutant information set are integrated, which may include matching different data sets according to time or place for subsequent analysis. The integrated data set is analyzed to determine the primary pollutants, secondary pollutants and the interactions and changes between them under specific time, place and meteorological conditions. The composite pollutant information obtained by analysis is integrated to construct a composite pollutant information data set, which contains detailed information such as the types, concentrations, spatiotemporal distribution of composite pollutants, and their relationship with meteorological parameters and primary pollutants.

[0069] The analysis methods of the chemical reactivity information set include: C1: Based on the chemical composition of each pollutant gas in the primary pollutant information set and the chemical composition contained in the air in the air basic information data set, the chemical composition information set is integrated to obtain the chemical composition information set, and the structure and electronic configuration of each chemical component are obtained based on the comparison between the chemical composition information set and the chemical knowledge information data set, so as to obtain the potential reaction information set, C2: Based on the comparison between the chemical composition information set and the chemical knowledge information data set, the chemical properties and reaction conditions of each chemical component in the chemical composition information set are obtained, so as to obtain the chemical reaction type information set, C3: Based on the comparison between the meteorological parameter information set and the reaction conditions of each chemical component, the meteorological parameters that affect the reaction rate and occurrence probability are obtained, so as to obtain the target meteorological parameter set, C4: Integrate C1-C3 to obtain the chemical reactivity information set.

[0070] It should be noted that the chemical composition of each pollutant gas is extracted from the primary pollutant information set, and the chemical composition contained in the air is obtained from the air basic information data set. The above two parts of chemical composition information are integrated to form a complete chemical composition information set, and the integrated chemical composition information set is compared with the chemical knowledge information data set. The chemical knowledge information data set contains known chemical reaction mechanisms, structural formulas, and electronic configurations. Through comparison, the structure and electronic configuration of each chemical component are determined, so as to predict the chemical reactions that may occur between them and form a potential reaction information set. Based on the comparison between the chemical composition information set and the chemical knowledge information data set, each chemical component is analyzed. chemical properties and reaction conditions, determine the possible chemical reaction types, such as oxidation, reduction, addition, substitution, etc., according to the chemical properties and reaction conditions, organize these chemical reaction types into a chemical reaction type information set, obtain meteorological parameter information sets from meteorological monitoring stations, including temperature, humidity, wind speed, wind direction, air pressure, etc., analyze the effects of meteorological parameters on chemical reaction rates and occurrence probabilities, screen out meteorological parameters that have a significant impact on the reactions, form a target meteorological parameter set, integrate the potential reaction information set, chemical reaction type information set and target meteorological parameter set, comprehensively consider potential reactions, chemical reaction types and meteorological parameters that affect reactions, and form a complete chemical reactivity information set.

[0071] The method for extracting the production condition information set includes: D1: based on each composite pollutant in the composite pollutant information data set, identifying the key conditions required for the production of each composite pollutant, and obtaining the key condition information set; D2: based on the key condition information set, classifying and sorting according to meteorological parameters, pollutant types, concentrations and reaction times to obtain an integrated information set; D3: finally obtaining the production condition information set after verification based on the integrated information set.

[0072] It should be noted that the types and related information of each composite pollutant are identified from the composite pollutant information data set. For each composite pollutant, the key conditions required for its production are analyzed. These key conditions may include specific meteorological parameters (such as temperature range, humidity range, wind speed threshold, etc.), pollutant type and concentration threshold, reaction time, etc. The identified key condition information is integrated to construct a key condition information set. This information set contains the key conditions required for the production of each composite pollutant. According to the dimensions of meteorological parameters, pollutant type, concentration and reaction time, the key condition information set is classified and sorted, and similar conditions are classified or sorted according to the importance of the conditions. , integrate the classified key condition information to construct an integrated information set. This information set is clearer in structure and convenient for subsequent analysis and verification. The integrated information set is verified through actual monitoring data, experimental data or simulation data. The purpose of verification is to confirm whether the conditions in the integrated information set are true and reliable, and whether they can accurately reflect the conditions for the generation of complex pollutants. After verification, the reliable integrated information set is used as the generation condition information set. If certain conditions are found to be inaccurate or unreliable during the verification process, they can be corrected or deleted. By analyzing the generation condition information set, we can understand the generation of complex pollutants under different conditions and provide a scientific basis for environmental management and decision-making.

[0073] Methods for determining key locations: T1: Based on the basic information dataset and matching information set of the industrial park, GIS technology is used to determine the location of key emission sources, and based on the meteorological information dataset, meteorological software is used to determine the pollutant diffusion path and range. T2: Based on the basic information dataset of the industrial park, the pollutant diffusion path and range, GIS technology is used to determine the pollutant accumulation area, and based on the historical monitoring information dataset, GIS technology is used to determine the pollutant diffusion law and historical monitoring locations. T3: Based on the location of key emission sources, pollutant accumulation areas and historical monitoring locations, GIS technology is used to obtain the key location information set, thereby obtaining the key location.

[0074] It should be noted that the industrial park basic information dataset and matching information set should be imported into the GIS (geographic information system) software. Through the spatial analysis function of GIS, the locations of key emission sources can be identified and marked. These emission sources may be factories or facilities with large pollutant emissions. Meteorological software (such as meteorological simulation software) is used in combination with meteorological information datasets to simulate pollutant diffusion. The location and emission parameters of key emission sources (such as emission volume, emission height, emission rate, etc.) are input, meteorological conditions (such as wind speed, wind direction, temperature, etc.) are set, and the simulation program is run to obtain the diffusion path and range of pollutants. The industrial park basic information dataset and pollutant diffusion simulation results are imported into GIS. The spatial overlay analysis function of GIS is used to combine the terrain, landform, building layout and other information of the industrial park to determine the areas where pollutants are easily accumulated. The historical monitoring information dataset should be imported. The dataset should contain information such as the location of historical monitoring points, monitoring time, and monitoring results. The historical monitoring data is analyzed in GIS to understand the pollutants. The diffusion law of pollutants, such as diffusion speed and diffusion direction, is used to mark the location of historical monitoring points for reference and verification in subsequent analysis. The location of key emission sources, pollutant accumulation areas and historical monitoring locations are integrated in GIS. Based on the integrated information, the buffer analysis, overlay analysis and other functions of GIS are used to determine the key locations. These key locations may be near emission sources with large pollutant emissions, the center or edge of pollutant accumulation areas, and areas with high pollutant concentrations as shown by historical monitoring results. The determined key location information is organized into a key location information set, including location coordinates, type (such as emission source, accumulation area, monitoring point, etc.), related pollutants and other information. The determined key locations are verified and optimized according to actual conditions and needs. Verification can be carried out through field surveys, monitoring data comparisons and other methods to ensure the accuracy and reliability of the key locations. According to the verification results, the key location information set is adjusted and optimized as necessary to improve its application value in environmental management and decision-making.

[0075] The key monitoring methods include: Y1: obtaining the current meteorological parameters based on the meteorological information data set, matching the current meteorological parameters with the target meteorological parameter set to obtain the target meteorological parameters that are most similar to the current meteorological parameters, thereby obtaining the corresponding target meteorological parameter set, Y2: matching the target meteorological parameter set with the generation condition information set to obtain the corresponding generation condition information set, Y3: matching the corresponding generation condition information set with the composite pollutant information data set to obtain the corresponding composite pollutant information set, and then evaluating the characteristics of each composite pollutant in the corresponding composite pollutant information set to obtain an evaluation information set, and prioritizing each evaluation information in the evaluation information set, Y4: adjusting the monitoring frequency of the monitoring equipment through the remote adjustment software based on the sorted evaluation information set, increasing the monitoring frequency for high-priority composite pollutants, and reducing the monitoring frequency for low-priority composite pollutants.

[0076] It should be noted that the current meteorological parameters are obtained in real time from the meteorological information data set. These parameters may include temperature, humidity, wind speed, wind direction, atmospheric pressure, etc. The current meteorological parameters are matched with the target meteorological parameter set to find the target meteorological parameters closest to the current meteorological parameters. This can be achieved by calculating the similarity or difference between the meteorological parameters. For example, using methods such as Euclidean distance and cosine similarity, after determining the target meteorological parameters closest to the current meteorological parameters, the corresponding target meteorological parameter set is obtained. This set contains the key factors that may affect the generation and diffusion of pollutants under current meteorological conditions. After determining the target meteorological parameters closest to the current meteorological parameters, the corresponding target meteorological parameter set is obtained. This set contains the key factors that may affect the generation and diffusion of pollutants under current meteorological conditions. The obtained target meteorological parameter set is matched with the production condition information set. The production condition information set may contain different production processes or equipment under different meteorological conditions. The operating parameters, emission characteristics and other information of the plant are obtained. After successful matching, the generation condition information set corresponding to the current meteorological conditions and production process is obtained, and the obtained generation condition information set is used to match the composite pollutant information data set. The composite pollutant information data set contains information such as the generation conditions, characteristics, and degree of harm of various composite pollutants. The matched composite pollutant information is evaluated, and the composite pollutant information is evaluated according to the characteristics of the composite pollutants (such as toxicity, diffusivity, stability, etc.), and an evaluation information set is generated. The various evaluation information in the evaluation information set are prioritized and sorted according to factors such as the degree of harm of the pollutants and the possibility of diffusion under current meteorological conditions. According to the sorted evaluation information set, the monitoring frequency of the monitoring equipment is adjusted through the remote adjustment software. For high-priority composite pollutants, the monitoring frequency is increased to timely discover and deal with possible environmental pollution incidents. For low-priority composite pollutants, the monitoring frequency can be appropriately reduced to save resources and costs.

[0077] The processing module is connected to the control terminal, and the monitoring equipment, meteorological monitoring station, acquisition module, processing module and output module at the key position are connected through a proprietary LAN. The method for creating a proprietary LAN includes: N1: information acquisition, obtaining the basic networking information of the acquisition module, processing module, output module, monitoring equipment and meteorological monitoring station, the basic networking information includes the device IP, N2: device connection, the control terminal automatically allocates the communication channel to match the IP of the acquisition module, the IP of the processing module, the IP of the output module, the IP of the monitoring equipment and the IP of the meteorological monitoring station, N3: configuration networking, the control terminal configures the corresponding network services and security measures for the communication channel to create a proprietary LAN.

[0078] It should be noted that in a private LAN, the control terminal can configure a DHCP server to dynamically assign IP addresses to connected devices. In a private LAN, an internal DNS server can be configured to map device names to IP addresses and provide NTP, which is used to synchronize device time in the network to ensure time consistency. In environmental monitoring systems, time synchronization is essential for data analysis and historical records. Centralized storage and management of data, such as NAS (Network Attached Storage) or SAN (Storage Area Network), can be used to store monitoring data, configuration files, log files, etc. Remote management services are provided, and secure, encrypted remote connections are provided to facilitate remote maintenance and configuration by administrators. Security measures include firewalls, access control lists, encrypted communications, security audits and logging, user authentication and authorization, and security updates and patch management.

[0079] The monitoring equipment includes an air quality monitoring station, and the air quality monitoring station is equipped with a variety of gas analyzers that can monitor multiple pollutants simultaneously.

[0080] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is limited by the attached embodiments and their equivalents.

Claims

1. An industrial aggregated composite pollutant identification system based on big data, comprising a collection module, a processing module and an output module, characterized in that: The acquisition module stores the industrial park basic information data set, the chemical knowledge information data set, the factory emission information data set and the emission index information data set. The acquisition module sets monitoring equipment at key locations in the industrial park to regularly monitor the air quality of the industrial park to obtain a monitoring information data set and a monitoring equipment information data set. The acquisition module sets a meteorological monitoring station in the industrial park to monitor the weather in the industrial park to obtain a meteorological information data set. The industrial park basic information data set includes an air basic information data set, and the monitoring information data set includes a real-time monitoring information data set and a historical monitoring information data set. A prediction and analysis model is provided in the processing module. The processing module derives a composite pollutant information data set based on the factory emission information data set and the meteorological information data set in the acquisition module according to the pollutant derivation method, and extracts the generation conditions of the composite pollutants based on the composite pollutant information data set to obtain a generation condition information set. The processing module adjusts the monitoring frequency of the monitoring equipment based on the generation condition information set and the meteorological information data set in combination with the key monitoring method. The processing module matches and associates the factory emission information data set with the emission index information data set to obtain a matching information set. The prediction and analysis model obtains an air quality report based on the real-time monitoring information data set and the matching information set, and obtains a predicted composite pollutant report based on the air quality report and the meteorological information data set. Finally, the processing module obtains an emission-exceeding-standard factory information set by matching the air quality report with the matching information set. The output module outputs identification reports based on air quality reports, factories with excessive emissions, and forecasted composite pollutant reports; The predictive analysis model includes: S1: The industrial park basic information dataset, chemical knowledge information dataset, factory emission information dataset, emission index information dataset, meteorological information dataset and monitoring information dataset are preprocessed to obtain standard data, which are divided into training set, validation set and test set, and a database is set up to integrate and store the standard data; S2: Select the long short-term memory network as the feature extractor to extract the features of the training set and extract the target features in the training set; S3: Train the feature extractor based on the training set to obtain an initial model for analyzing the air quality of the industrial park and predicting the generation of composite pollutants; S4: Train and optimize the initial model to obtain a predictive analysis model, and regularly update the predictive analysis model; The pollutant derivation method includes: B1: Determine the type and chemical composition of a single pollutant based on the factory emission information data set to obtain the primary pollutant information set; B2: Determine wind speed, wind direction, temperature, humidity and atmospheric pressure based on the meteorological information data set, thereby obtaining a meteorological parameter information set; B3: The chemical reactivity information set is obtained by analyzing each pollutant gas in the primary pollutant information set, and then combined with the meteorological parameter information set and the chemical reaction equation in the chemical knowledge information data set for comparison and analysis to obtain the secondary pollutant information set; B4: A composite pollutant information dataset is obtained by integrating the primary pollutant information set, the meteorological parameter information set and the secondary pollutant information set; The analysis method of the chemical reactivity information set comprises: C1: The chemical composition information set is obtained by integrating the chemical composition of each pollutant gas in the primary pollutant information set and the chemical composition of the air in the air basic information data set. The structure and electronic configuration of each chemical component are obtained by comparing the chemical composition information set with the chemical knowledge information data set, thereby obtaining the potential reaction information set; C2: Based on the comparison between the chemical composition information set and the chemical knowledge information data set, the chemical properties and reaction conditions of each chemical component in the chemical composition information set are obtained, thereby obtaining the chemical reaction type information set; C3: Compare the meteorological parameter information set with the reaction conditions of each chemical component to obtain the meteorological parameters that affect the reaction rate and occurrence probability, thereby obtaining the target meteorological parameter set; C4: Integrate C1-C3 to obtain the chemical reactivity information set.

2. The industrial aggregated composite pollutant identification system based on big data according to claim 1 is characterized by: The training and optimization methods include: A1: Model evaluation: evaluate the trained initial model based on the validation set to obtain the evaluation results, which are used to analyze the performance of the model and identify existing problems and improvement points. A2: Model optimization: optimize the model based on the evaluation results and use the grid search method to tune the parameters; A3: Model verification: verify the optimized model based on the test set to ensure the generalization ability of the model on unknown data; A4: Model update: regularly update the predictive analysis model and use incremental learning methods to update the model in real time.

3. The industrial aggregated composite pollutant identification system based on big data according to claim 1 is characterized by: The method for extracting the generation condition information set includes: D1: Based on each composite pollutant in the composite pollutant information data set, identify the key conditions required for the generation of each composite pollutant and obtain the key condition information set; D2: Based on the key condition information set, the integrated information set is obtained by classifying and sorting according to meteorological parameters, pollutant types, concentrations and reaction times; D3: After verification based on the integrated information set, the generation condition information set is finally obtained.

4. The industrial aggregated composite pollutant identification system based on big data according to claim 1 is characterized by: Method for determining the key position: T1: Based on the industrial park basic information data set and matching information set, GIS technology is used to determine the location of key emission sources, and based on the meteorological information data set, meteorological software is used to determine the diffusion path and range of pollutants; T2: Based on the basic information data set of the industrial park, the diffusion path and range of pollutants, GIS technology is used to determine the pollutant accumulation area; based on the historical monitoring information data set, GIS technology is used to determine the pollutant diffusion law and historical monitoring location; T3: Based on the locations of key emission sources, pollutant accumulation areas and historical monitoring locations, GIS technology is used to obtain key location information sets to obtain key locations.

5. The industrial aggregated composite pollutant identification system based on big data according to claim 1 is characterized by: The key monitoring methods include: Y1: Based on the meteorological information data set, the current meteorological parameters are obtained, and the target meteorological parameters that are most similar to the current meteorological parameters are matched with the target meteorological parameter set, so as to obtain the corresponding target meteorological parameter set; Y2: Acquire the corresponding generation condition information set based on matching the target meteorological parameter set with the generation condition information set; Y3: Based on the matching of the corresponding generation condition information set and the composite pollutant information data set, the corresponding composite pollutant information set is obtained, and then the characteristics of each composite pollutant in the corresponding composite pollutant information set are evaluated to obtain an evaluation information set, and each evaluation information in the evaluation information set is prioritized; Y4: Based on the sorted evaluation information set, the monitoring frequency of the monitoring equipment is adjusted through the remote adjustment software to increase the monitoring frequency for high-priority composite pollutants and reduce the monitoring frequency for low-priority composite pollutants.

6. The industrial aggregated composite pollutant identification system based on big data according to claim 1 is characterized by: The processing module is connected to a control terminal, and the monitoring equipment, meteorological monitoring station, acquisition module, processing module and output module at key positions are all connected through a proprietary local area network. The method for creating the proprietary local area network includes: N1: Information acquisition, obtaining basic networking information of the acquisition module, processing module, output module, monitoring equipment and meteorological monitoring station, the basic networking information including the equipment IP; N2: Device connection, the control terminal automatically allocates the communication channel to match the IP of the acquisition module, the IP of the processing module, the IP of the output module, the IP of the monitoring device and the IP of the meteorological monitoring station; N3: Configure the network, control the terminal to configure the corresponding network services and security measures for the communication channel, and create a private LAN.

7. The industrial aggregated composite pollutant identification system based on big data according to claim 1 is characterized by: The monitoring equipment includes an air quality monitoring station, and the air quality monitoring station is equipped with multiple gas analyzers and can monitor multiple pollutants simultaneously.

Images