Glass pollutant real-time monitoring and early warning method and system

Abstract

The invention discloses a real-time monitoring and early warning method and system for glass pollutants, and the method can more accurately monitor and analyze various pollutants generated in a glass production process through the establishment of a multi-factor comprehensive analysis model and the combination of big data analysis and a machine learning algorithm, reduces the misinformation and missing report conditions, and improves the production efficiency. Advanced technical means and algorithm optimization are utilized, especially factors such as the operation state of production equipment, the quality of raw materials and meteorological conditions are comprehensively considered, and the prediction capacity of the pollutant concentration change trend is improved. Therefore, an enterprise can send out an accurate early warning signal in advance, and a reliable basis is provided for taking timely and effective countermeasures.

Description

Technical Field

[0001] The present invention relates to the field of glass production, and particularly to a method and system for real-time monitoring and early warning of glass pollutants. Background Art

[0002] A large amount of pollutants will be emitted during the glass manufacturing process. A large amount of pollutants will be generated during the glass production process, such as particulate matter, sulfur dioxide, nitrogen oxides, heavy metals, etc. These pollutants not only cause serious harm to the environment, but also affect the health of surrounding residents.

[0003] Existing early warning models may be established based on historical data, and are insufficiently adaptable to changes in glass production processes or new pollution situations. If an enterprise adopts new raw materials or production processes, it may lead to changes in pollutant emission characteristics, and existing early warning models may not be able to accurately predict such changes; the accuracy of the model may be affected by data quality. If there are errors or omissions in the monitoring data, it may lead to inaccurate output results of the early warning model. The setting of the warning threshold may be too single or fixed, lacking flexibility. For example, for different production stages, different pollutant types, or different environmental conditions, different warning thresholds may be required, but existing systems may not be able to automatically adjust the threshold. The setting of the warning threshold lacks a scientific basis and is only based on experience or regulatory requirements, which may lead to false alarms or missed alarms.

[0004] In summary, a method and system for real-time monitoring and early warning of glass pollutants are needed to solve the deficiencies in the prior art. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention provides a method and system for real-time monitoring and early warning of glass pollutants, aiming to solve the above problems.

[0006] To achieve the above object, the present invention provides the following technical solution: A method for real-time monitoring and early warning of glass pollutants, comprising the following steps: Step S1: Data collection, collecting environmental protection data during the glass production process; Step S2: Data preprocessing, cleaning and organizing the collected data, identifying and removing outliers and missing data, and normalizing or standardizing the data; Step S3: Establishing a multi-factor comprehensive analysis model, constructing a comprehensive evaluation index model based on big data analysis and machine learning algorithms, and predicting the change trend of pollutant concentration; Step S4: Warning threshold setting and dynamic adjustment, determining the initial warning threshold in combination with historical data and experience, and automatically adjusting the warning threshold through real-time data analysis and production process change conditions; Step S5: Model and Application. The model is verified using cross-validation and prediction error analysis methods. The optimized model is applied to actual production, and production data is collected in real time.

[0007] Optionally, the multi-factor comprehensive analysis model in step S3 is established in the following manner: Step A1: Determine the research objective and evaluation indicators, clarify the objective to be optimized, and determine the corresponding comprehensive evaluation indicators; Step A2: Collect data, and collect data and the corresponding comprehensive evaluation indicator values through experiments and production records; Step A3: Preprocess the collected data to remove outliers and incorrect data; Step A4: Determine the model form, and select a suitable model form based on theoretical analysis and experience; Step A5: Use the data fitting method to determine the parameter values in the model; Step A6: Verify the model using cross-validation and prediction error analysis.

[0008] Optionally, the comprehensive evaluation indicator value in step A2 is calculated in the following manner: F = f(T, P, R1, R2, R3…, t) = , where T represents temperature, P represents pressure, R1, R2, R3… represent the proportion of different raw material components, t represents production time, a1, a2, a3, a4 respectively represent the initial influence degrees in different situations, b1, b2, e1, f1 represent the power exponents of temperature, log(T) represents the logarithmic term of temperature, and c1 represents the power exponent of pressure.

[0009] Optionally, the comprehensive evaluation indicator value in step A2 is calculated in the following manner: F = f(T, P, R1, R2, R3…, t) = + + + , where T represents temperature, P represents pressure, R represents the proportion of different raw material components, t represents production time, a1; a2; a3; a4 represent the initial influence degrees in different situations, b1; b2; represent the power exponents of temperature, c1; ; represent the power exponent of pressure, and b3 represents the power exponent of production time, represents the power exponent in the combined action of temperature with production time, pressure, and raw material composition ratio, ; represents the logarithmic term of temperature, ; represents the logarithmic term of pressure, represents the raw material composition ratio R z1 's power exponent, ; ; ; represents the logarithmic term of the raw material composition ratio, is the power exponent of different raw material composition ratios R i ; represents the power exponent of the raw material composition ratio R k ; represents the power exponent of the raw material composition ratio R u ;

[0010] Optionally, the method further includes collecting real-time pollutant monitoring data during the glass production process through Internet of Things technology, transmitting the data to the central processing system, processing and analyzing the collected data using big data analysis technology to generate a pollutant concentration change trend graph, predicting the pollutant concentration change trend through machine learning algorithms, and automatically sending a warning signal when the pollutant concentration exceeds the warning threshold.

[0011] Optionally, the method further includes providing a visual interface through the visual interface module to display the pollutant concentration data and warning information in real time.

[0012] Optionally, the method further includes automatically adjusting the glass production process parameters according to the warning signal, and optimizing the model parameters according to the latest production data and environmental conditions.

[0013] Optionally, the method further includes automatically adjusting the warning threshold according to the real-time monitoring data and changes in the production process. When the production process changes or the environmental conditions change, the system can automatically adjust the warning threshold.

[0014] A real-time monitoring and warning system for glass pollutants, adopting the real-time monitoring and warning method for glass pollutants, includes a data collection module, a data processing module, a warning model module, a warning output module, and a visual interface module, The data collection module is used to collect real-time pollutant monitoring data during the glass production process; The data processing module is used to clean, preprocess, and normalize the collected data; The warning model module is used to establish and optimize a multi-factor comprehensive analysis model, and dynamically adjust the warning threshold according to real-time data; An early warning output module, which is used to send out an early warning signal when the pollutant concentration exceeds the early warning threshold; A visualization interface module, which is used to display the pollutant concentration data and early warning information in real time.

[0015] Optionally, the system further includes an automatic control module and a data storage module, An automatic control module, which is used to automatically adjust the glass production process parameters according to the early warning signal; A data storage module, which is used to store historical monitoring data and early warning records.

[0016] Advantages of the present invention: 1. In the present invention, by establishing a multi-factor comprehensive analysis model and combining big data analysis and machine learning algorithms, this method can more accurately monitor and analyze various pollutants generated in the glass production process, reducing false alarms and missed reports; 2. In the present invention, by using advanced technical means and algorithm optimization, especially by comprehensively considering factors such as the operating status of production equipment, the quality of raw materials, and meteorological conditions, the prediction ability of the pollutant concentration change trend is improved. This enables enterprises to send out accurate early warning signals in advance, providing a reliable basis for taking timely and effective countermeasures; 3. In the present invention, the system can realize the dynamic adjustment of the early warning threshold, automatically adjust the threshold according to the real-time data and the changes in the production process, thereby improving the accuracy and timeliness of the early warning. This method overcomes the problem of lack of flexibility of the traditional fixed early warning threshold, ensuring that scientific and reasonable early warning standards can be set in different production stages or environmental conditions.

[0017] 4. In the present invention, through the Internet of Things technology, the pollutant monitoring data in the production process is collected in real time, and processed and analyzed by using big data analysis technology and machine learning algorithms to generate a pollutant concentration change trend chart, which not only helps enterprises optimize the production process parameters, but also can continuously optimize the model parameters according to the latest production data and environmental conditions. Description of the Drawings

[0018] Figure 1 It is a schematic flow chart of a method of the present invention.

[0019] Figure 2 It is a schematic structural diagram of a system of the present invention. Detailed Embodiments

[0020] To more clearly illustrate the technical solutions in the embodiments of the invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0021] As Figure 1 shown, a real-time monitoring and early warning method for glass pollutants includes the following steps: Step S1: Data collection, collecting environmental protection data during the glass production process; Step S2: Data preprocessing, cleaning and organizing the collected data, identifying and removing outliers and multi-source data, and performing normalization or standardization processing on the data; Step S3: Establishing a multi-factor comprehensive analysis model, based on big data analysis and machine learning algorithms, constructing a comprehensive evaluation index model to predict the change trend of pollutant concentration; Step S4: Warning threshold setting and dynamic adjustment, determining the initial warning threshold in combination with historical data and experience, and automatically adjusting the warning threshold through real-time data analysis and production process changes; Step S5: Model and application, verifying the model using cross-validation and prediction error analysis methods, applying the optimized model to actual production, and collecting production data in real time.

[0022] Optionally, the multi-factor comprehensive analysis model in step S3 is established in the following manner: Step A1: Determining the research objective and evaluation index, clarifying the objective to be optimized, and determining the corresponding comprehensive evaluation index; Step A2: Collecting data, collecting data and the corresponding comprehensive evaluation index values through experiments and production records; Step A3: Preprocessing the collected data, removing outliers and incorrect data; Step A4: Determining the model form, selecting a suitable model form based on theoretical analysis and experience; Step A5: Using the data fitting method to determine the parameter values in the model; Step A6: Verifying the model using cross-validation and prediction error analysis.

[0023] The method also includes collecting real-time pollutant monitoring data during the glass production process through Internet of Things technology, transmitting the data to a central processing system, processing and analyzing the collected data using big data analysis technology to generate a pollutant concentration change trend chart, predicting the pollutant concentration change trend through machine learning algorithms, and automatically sending out a warning signal when the pollutant concentration exceeds the warning threshold; providing a visual interface through a visual interface module to display the pollutant concentration data and warning information in real time; automatically adjusting the glass production process parameters according to the warning signal, and optimizing the model parameters according to the latest production data and environmental conditions; automatically adjusting the warning threshold according to the real-time monitoring data and changes in the production process, and the system can automatically adjust the warning threshold when the production process changes or the environmental conditions change.

[0024] Example 1: The first calculation method, the comprehensive evaluation index value in step A2 is calculated as follows: F = f(T, P, R1, R2, R3…, t) = , In the formula, T represents temperature, P represents pressure, R1, R2, R3… represent the proportion of different raw material components, t represents the production time, a1, a2, a3, a4 respectively represent the initial influence degree in different situations, b1, b2, e1, f1 represent the power exponents of temperature, log(T) represents the logarithmic term of temperature, and c1 represents the power exponent of pressure.

[0025] Example 2: The second calculation method, the comprehensive evaluation index value in step A2 is calculated as follows: F = f(T, P, R1, R2, R3…, t) = + + + , In the formula, T represents temperature, P represents pressure, R represents the proportion of different raw material components, t represents the production time, a1; a2; a3; a4 represent the initial influence degree in different situations, b1; b2; represent the power exponents of temperature, c1; ; represent the power exponent of pressure, b3 represents the power exponent of birth time, represents the power exponent of temperature in the combined action with production time, pressure and raw material component proportion, ; represents the logarithmic term of temperature, ; represents the logarithmic term of pressure, represents the raw material component proportion Rz1 Power exponent, ; ; ; Logarithmic term representing the proportion of raw material components, is the power exponent of different raw material component ratios R i Power exponent, represents the power exponent of raw material component ratio R k Power exponent, represents the power exponent of raw material component ratio R u Power exponent.

[0026] In this term, a1 reflects the initial influence degree of the combined action of temperature, pressure and raw material component ratio on the comprehensive evaluation index. b1 is the power exponent of temperature, which determines the sensitivity of the comprehensive evaluation index to temperature changes. c1 is the power exponent of pressure, reflecting the influence degree of pressure in this combined action, is the power exponent of different raw material component ratios R i reflecting the influence degree of each raw material component on the comprehensive evaluation index under the combined action with temperature and pressure. Introducing the logarithmic term of the raw material component ratio further refines the influence of the raw material component ratio on the comprehensive evaluation index.

[0027] In this term, a2 also represents an initial influence degree, b2 is another power exponent of temperature, reflecting different temperature influence modes, is the power exponent of pressure, and different j values correspond to different pressure influence modes. Introducing the logarithmic term of pressure refines the influence of pressure on the comprehensive evaluation index, represents the power exponent of raw material component ratio R k reflecting the influence of different raw material components under another temperature and pressure action mode. The logarithmic term enriches the expression of the influence of the raw material component ratio.

[0028] In this term, a3 is the initial influence degree related to the production time, b3 is the power exponent of the production time, reflecting the influence degree and non-linear relationship of the production time on the comprehensive evaluation index. is the power exponent of temperature in the combined action with production time, pressure and raw material component ratio. Introducing the logarithmic term of temperature further refines the influence of temperature on the comprehensive evaluation index, is the power exponent of pressure, acting together with production time, temperature and raw material component ratio. The logarithmic term further reflects the complex influence of pressure. In this term, Represents the proportion R of raw material components u The power exponent, which acts together with production time, temperature, and pressure. The logarithmic term of further reflects the complex influence of the proportion of raw material components.

[0029] In In this term, a4 is another degree of comprehensive influence, which is another power exponent of temperature, representing different ways of temperature influence. The logarithmic term of also enriches the expression of the influence of temperature, z1 is the power exponent of the raw material component ratio R z1 which plays an influence in the complex interaction. The logarithmic term of enriches the expression of the influence of the raw material component ratio, which is the power exponent of production time, representing the way of production time's influence in this complex interaction.

[0030] This application establishes a multi-factor comprehensive analysis model, fully considering the interaction between multiple factors in the glass production process. For example, factors such as the operating state of production equipment, the quality of raw materials, and meteorological conditions are incorporated into the model. Through big data analysis and machine learning algorithms, a more accurate early warning model is established, and a scientific and reasonable method for setting early warning thresholds is established. Combining historical data, regulatory requirements, and expert experience, the initial early warning threshold is determined. By statistically analyzing historical data, the distribution range and outliers of pollutant concentrations are determined, and reasonable early warning thresholds are set in combination with regulatory requirements and expert experience.

[0031] Realize the dynamic adjustment of the early warning threshold, automatically adjusting the threshold according to real-time data and production conditions. For example, when the production process changes or the environmental conditions change, the system can automatically adjust the early warning threshold to improve the accuracy and timeliness of early warning.

[0032] Clean the collected environmental protection data to remove outliers and incorrect data. Statistical methods such as the 3-sigma rule or the box plot method can be used to identify outliers and correct or delete them according to specific situations. At the same time, the data is normalized or standardized for subsequent analysis and modeling. The multi-factor comprehensive analysis model is mainly used. The multi-factor comprehensive analysis model can help us better understand the interaction between various factors, thereby optimizing the production process.

[0033] As Figure 2 shown, a real-time monitoring and early warning system for glass pollutants, adopting the real-time monitoring and early warning method for glass pollutants, includes a data acquisition module, a data processing module, an early warning model module, an early warning output module, and a visualization interface module. The data acquisition module is used to collect real-time pollutant monitoring data during the glass production process; A data processing module for cleaning, preprocessing, and normalizing the collected data; An early warning model module for establishing and optimizing a multi-factor comprehensive analysis model and dynamically adjusting the early warning threshold according to real-time data; An early warning output module for sending out an early warning signal when the pollutant concentration exceeds the early warning threshold; A visualization interface module for real-time displaying of pollutant concentration data and early warning information.

[0034] The system further includes an automatic control module and a data storage module. An automatic control module for automatically adjusting the glass production process parameters according to the early warning signal; A data storage module for storing historical monitoring data and early warning records.

[0035] By establishing a multi-factor comprehensive analysis model and combining big data analysis and machine learning algorithms, the method of the present invention can more accurately monitor and analyze various pollutants (such as sulfur dioxide, nitrogen oxides, particulate matter, etc.) generated during the glass production process, reduce false alarms and missed alarms, and improve the prediction ability of the pollutant concentration change trend by using advanced technical means and algorithm optimization, especially by comprehensively considering factors such as the operating state of production equipment, the quality of raw materials, and meteorological conditions. This enables enterprises to issue accurate early warning signals in advance and provides a reliable basis for taking timely and effective countermeasures.

[0036] The system can dynamically adjust the early warning threshold, automatically adjusting the threshold according to real-time data and changes in the production process, thereby improving the accuracy and timeliness of early warning. This method overcomes the problem of lack of flexibility of traditional fixed early warning thresholds, ensuring that scientific and reasonable early warning standards can be set in different production stages or environmental conditions. By using Internet of Things technology to collect real-time pollutant monitoring data during the production process and using big data analysis technology and machine learning algorithms for processing and analysis, a pollutant concentration change trend graph is generated. This not only helps enterprises optimize production process parameters but also continuously optimizes model parameters according to the latest production data and environmental conditions.

[0037] A visualization interface module is provided for real-time displaying of pollutant concentration data and early warning information, facilitating enterprise managers to intuitively understand the pollution situation during the production process and make rapid responses. The system helps enterprises achieve effective control of pollutant emissions.

[0038] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, or improvements made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A real-time monitoring and early warning method for glass pollutants, characterized in that, It includes the following steps: Step S1: Data collection, collecting environmental protection data during the glass production process; Step S2: Data preprocessing, cleaning and sorting the collected data, identifying and removing outliers and multivariate data, and performing normalization or standardization processing on the data; Step S3: Establishing a multi-factor comprehensive analysis model, based on big data analysis and machine learning algorithms, constructing a comprehensive evaluation index model to predict the changing trend of pollutant concentration; Step S4: Early warning threshold setting and dynamic adjustment, determining the initial early warning threshold in combination with historical data and experience, and automatically adjusting the early warning threshold through real-time data analysis and production process changes; Step S5: Model and application, verifying the model using cross-validation and prediction error analysis methods, applying the optimized model to actual production, and collecting production data in real time.

2. The real-time monitoring and early warning method for glass pollutants according to claim 1, wherein, The multi-factor comprehensive analysis model in step S3 is established in the following way: Step A1: Determine the research objective and evaluation index, clarify the objective to be optimized, and determine the corresponding comprehensive evaluation index; Step A2: Collect data, collect data and the corresponding comprehensive evaluation index values through experiments and production records; Step A3: Preprocess the collected data, removing outliers and incorrect data; Step A4: Determine the model form, select a suitable model form based on theoretical analysis and experience; Step A5: Use the data fitting method to determine the parameter values in the model; Step A6: Verify the model using cross-validation and prediction error analysis.

3. The real-time monitoring and early warning method for glass pollutants according to claim 2, wherein The comprehensive evaluation index value in step A2 is calculated in the following way: F = f(T, P, R1, R2, R3…, t) = ， In the formula, T represents temperature, P represents pressure, R1, R2, R3… represent the proportion of different raw material components, t represents production time, a1, a2, a3, a4 respectively represent the initial influence degree in different situations, b1, b2, e1, f1 represent the power exponents of temperature, log(T) represents the logarithmic term of temperature, and c1 represents the power exponent of pressure.

4. The real-time monitoring and early warning method for glass pollutants according to claim 1, wherein, The comprehensive evaluation index value in step A2 is calculated in the following way: F = f(T, P, R1, R2, R3…, t) = + + + ， Wherein, T represents temperature, P represents pressure, R represents the proportion of different raw material components, t represents the production time, a1, a2, a3, a4 represent the initial influence degrees under different conditions, b1, b2 represent the power exponent of temperature, c1 ; represent the power exponent of pressure, b3 represents the power exponent of the production time, represent the power exponent of temperature in the combined action with the production time, pressure and proportion of raw material components, ; represent the logarithmic term of temperature, ; represent the logarithmic term of pressure, represents the proportion of raw material components R z1 of the power exponent, ; ; ; represent the logarithmic term of the proportion of raw material components, is the power exponent of different proportions of raw material components R i of, represents the power exponent of the proportion of raw material components R k of, represents the power exponent of the proportion of raw material components R u of the power exponent.

5. The real-time monitoring and early warning method for glass pollutants according to claim 1, characterized in that, The method also includes using Internet of Things technology to collect pollutant monitoring data during the glass production process in real time, transmitting the data to the central processing system, using big data analysis technology to process and analyze the collected data, generating a changing trend chart of pollutant concentration, predicting the changing trend of pollutant concentration through machine learning algorithms, and automatically sending out a warning signal when the pollutant concentration exceeds the warning threshold.

6. The real-time monitoring and early warning method for glass pollutants according to claim 1, characterized in that The method also includes providing a visual interface through the visual interface module to display pollutant concentration data and warning information in real time.

7. The real-time monitoring and early warning method for glass pollutants according to claim 1, characterized in that, The method also includes automatically adjusting the glass production process parameters according to the warning signal, and optimizing the model parameters according to the latest production data and environmental conditions.

8. The real-time monitoring and early warning method for glass pollutants according to claim 1, characterized in that The method also includes automatically adjusting the warning threshold according to the real-time monitoring data and production process changes. When the production process changes or the environmental conditions change, the system can automatically adjust the warning threshold.

9. A real-time monitoring and early warning system for glass pollutants, which adopts the real-time monitoring and early warning method for glass pollutants according to any one of claims 1-8, is characterized in that, It includes a data collection module, a data processing module, a warning model module, a warning output module and a visual interface module. A data acquisition module for real-time collection of pollutant monitoring data during the glass production process; A data processing module for cleaning, preprocessing, and normalizing the collected data; An early warning model module for establishing and optimizing a multi-factor comprehensive analysis model and dynamically adjusting the early warning threshold according to real-time data; An early warning output module for sending out an early warning signal when the pollutant concentration exceeds the early warning threshold; A visualization interface module for real-time display of pollutant concentration data and early warning information.

10. The real-time monitoring and early warning system for glass pollutants according to claim 9, characterized in that, The system further includes an automatic control module and a data storage module, An automatic control module for automatically adjusting the glass production process parameters according to the early warning signal; A data storage module for storing historical monitoring data and early warning records.

Images