Intelligent control method of gas flow organization based on air pollution detection

Abstract

The present invention discloses an intelligent control method for gas flow organization based on air pollution detection, which relates to the technical field of air pollution detection. The method divides a laboratory into multiple monitoring areas and arranges air quality sensors to obtain pollutant concentration data in real time, extracts diffusion rate anomaly characteristics and peak fluctuation characteristics using pre-processed data, calculates the comprehensive pollution index of each area, and divides the area into different pollution degree levels according to the comprehensive pollution index and the set gradient standard threshold value, and adjusts the opening number of corresponding air vents and the air volume output strategy. The system also analyzes pollution changes within a fixed time period and dynamically adjusts ventilation parameters to ensure continuous optimization of indoor air quality. Through intelligent and dynamic monitoring and regulation, the instability and diffusivity of pollutants can be effectively dealt with, thereby achieving accurate and efficient air quality management.

Description

Technical Field

[0001] The present invention relates to the technical field of air pollution detection, and in particular to an intelligent control method for gas flow organization based on air pollution detection. Background Art

[0002] Air pollution detection refers to the process of monitoring and analyzing pollutants in the air using various technologies and equipment to evaluate air quality. However, in the design of ventilation systems in modern buildings, the traditional layout of vents and exhaust vents is often fixed and may not be adjusted according to the real-time changes in the amount of pollutant release and diffusion position, resulting in the system being unable to flexibly respond to changes in the distribution of pollutants, making it impossible to discharge pollutants in a timely and effective manner, especially in environments where pollutant generation is characterized by instability and diffusion, such as laboratories. At the same time, when pollutants diffuse at different speeds and directions in space, if the fixed ventilation system cannot form an effective airflow path to quickly discharge these pollutants, it may lead to a decline in local air quality. Summary of the invention

[0003] The purpose of the present invention is to provide a method for intelligent control of gas flow organization based on air pollution detection to solve the shortcomings of the background technology.

[0004] In order to achieve the above object, the present invention provides the following technical solution: a method for intelligent control of gas flow organization based on air pollution detection, comprising the following steps:

[0005] S1: Divide the laboratory into several monitoring areas, and arrange W air quality sensors in each monitoring area to obtain indoor pollutant concentration data in real time in different time periods;

[0006] S2: preprocessing the acquired indoor pollutant concentration data, and extracting the pollutant concentration diffusion rate abnormality characteristics and pollutant concentration peak fluctuation characteristics of the preprocessed indoor pollutant concentration data;

[0007] S3: According to the extracted abnormal characteristics of pollutant concentration diffusion rate and peak fluctuation characteristics of pollutant concentration, the weight assignment of indoor pollutant concentration data in each time period of the monitoring area is determined, and the comprehensive pollution index of each monitoring area is calculated after weighted average calculation;

[0008] S4: Compare and analyze the comprehensive pollution index of the monitoring area with the gradient standard threshold, divide the monitoring area into different pollution levels according to the analysis results, and determine the opening number and air volume output strategy of the corresponding air outlets in each monitoring area according to the division results;

[0009] S5: Further analyze the changes in pollution levels in each monitoring area within a fixed time period, and dynamically adjust the opening time and air volume output of the air vents based on the analysis results.

[0010] Preferably, in S2, the pollutant concentration diffusion rate anomaly characteristics of the pre-processed indoor pollutant concentration data are extracted and the pollutant concentration diffusion rate anomaly index is generated, and the method for obtaining the pollutant concentration diffusion rate anomaly index is:

[0011] Obtain pollutant concentration data at time t in real time and mark it as , and the pollutant diffusion rate at time t , the calculation expression is; ; and establish the diffusion wind speed data set at time t ; The acquired pollutant concentration data, pollutant diffusion rate and diffusion wind speed data constitute the nodes in the Bayesian network, and the structure diagram of the Bayesian network is constructed, which is expressed as: ; Construct the conditional probability table for each node, the diffusion rate Conditional probability table , using historical data, the parameters in the conditional probability table are estimated, and the parameter estimation expression is:

[0012] ;

[0013] In the formula, In historical data The number of simultaneous occurrences, express The number of simultaneous occurrences; for the current time t, the predicted diffusion rate distribution is calculated using the Bayesian network ; Given the actual observed diffusion rate , calculate its probability in the predicted distribution ; Calculate the pollutant concentration diffusion rate anomaly index, the expression is: ; In the formula, It is the pollutant concentration diffusion rate anomaly index.

[0014] Preferably, the pollutant concentration peak fluctuation characteristics of the pre-processed indoor pollutant concentration data are extracted and a pollutant concentration peak fluctuation index is generated. The method for obtaining the pollutant concentration peak fluctuation index is:

[0015] Obtain the pollutant concentration data C(s) within the time period M, where s represents the time point in the time series. Apply discrete Fourier transform to the pollutant concentration data C(s) in the time domain and convert it into frequency domain representation C(f). The expression is:

[0016] ;

[0017] In the formula, is the pollutant concentration data spectrum value at frequency f, N is the total number of signal sample points, is the kernel function of Fourier transform, j is the imaginary unit, and the amplitude of each frequency component in the frequency domain is calculated and marked as , the expression is: ;in, is the real part of C(f), is the imaginary part of C(f), in the calculated spectrum In the process, the main frequency component is identified, that is, the frequency point whose amplitude is greater than the preset standard amplitude. , calculate its energy , the expression is ; k is the number of the main frequency, and the energy of all main frequency components is weighted and summed to generate the pollutant concentration peak fluctuation index, which is expressed as:

[0018] ;

[0019] is the weight of each frequency component, It is the peak fluctuation index of pollutant concentration.

[0020] Preferably, in S3, the pollutant concentration diffusion rate anomaly index and the pollutant concentration peak fluctuation index are converted into a first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the weighted assignment labels of the indoor pollutant concentration data in each time period of the monitoring area as the prediction target, and takes minimizing the sum of the prediction errors of the weighted assignment labels of the indoor pollutant concentration data in all time periods as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The weight assignment of the indoor pollutant concentration data in each time period of the monitoring area is determined according to the model output results, and the weighted average calculation is performed on them to calculate the comprehensive pollution index of each monitoring area, wherein the machine learning model is a polynomial regression model.

[0021] Preferably, in S4, the comprehensive pollution index of the monitoring area is compared and analyzed with the gradient standard threshold, and the monitoring area is divided into different pollution degree levels according to the analysis results;

[0022] The obtained comprehensive pollution index of each monitoring area is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the comprehensive pollution index of each monitoring area is compared with the first standard threshold and the second standard threshold respectively;

[0023] If the comprehensive pollution index of each monitoring area is greater than the second standard threshold, the monitoring area will be classified as a high pollution level; if the comprehensive pollution index of each monitoring area is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the monitoring area will be classified as a medium pollution level; if the comprehensive pollution index of each monitoring area is less than the first standard threshold, the monitoring area will be classified as a low pollution level.

[0024] Preferably, in S5, the pollution level change of each monitoring area within a fixed time period is further analyzed, and the opening time and air volume output of the air vents are dynamically adjusted according to the analysis results;

[0025] According to the number of air vents opened and the air volume output strategy of each monitoring area, the ventilation system can generate a preliminary airflow path to discharge pollutants, and further analyze the changes in the pollution level of each monitoring area in the subsequent fixed time period. The comprehensive pollution index in each monitoring area that is greater than or equal to the first standard threshold and less than or equal to the second standard threshold is collected, a data set is established, and the mean and standard deviation of the data set are calculated. After analyzing the data, the time to open the air vents and the air volume output are dynamically adjusted according to the analysis results.

[0026] Preferably, if the mean value of the comprehensive pollution index in the data set is greater than or equal to the reference threshold value of the comprehensive pollution index mean value, and the standard deviation of the comprehensive pollution index is less than the reference threshold value of the comprehensive pollution index standard deviation, it indicates that the pollution source is stable but the pollution level is serious, and the opening time of the air outlet is extended, and the air volume output is increased to quickly reduce the pollutant concentration;

[0027] If the mean value of the comprehensive pollution index is greater than or equal to the reference threshold value of the mean value of the comprehensive pollution index, and the standard deviation of the comprehensive pollution index is greater than or equal to the reference threshold value of the standard deviation of the comprehensive pollution index, it indicates that there is an unstable and strong pollution source. Further extend the opening time of the air outlet, maximize the air volume output, and closely monitor the pollution changes;

[0028] If the mean value of the comprehensive pollution index is less than the reference threshold value of the mean value of the comprehensive pollution index, and the standard deviation of the comprehensive pollution index is greater than or equal to the reference threshold value of the standard deviation of the comprehensive pollution index, it indicates that the pollution source is unstable but the average concentration is low. Keep the current air outlet opening time and the existing air volume, and continue to monitor the fluctuation;

[0029] If the mean of the comprehensive pollution index is less than the reference threshold of the mean of the comprehensive pollution index, and the standard deviation of the comprehensive pollution index is less than the reference threshold of the standard deviation of the comprehensive pollution index, the air quality is good and stable, shorten the air vent opening time, and reduce the air volume output.

[0030] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0031] 1. The present invention obtains pollutant concentration data in real time by dividing the laboratory into multiple monitoring areas and arranging air quality sensors, and combines the abnormal characteristics of diffusion rate and peak fluctuation characteristics to accurately calculate the comprehensive pollution index of each area. According to the comparative analysis of the comprehensive pollution index and the gradient standard threshold, the opening time of the air vent and the air volume output strategy are dynamically adjusted to achieve differentiated treatment of areas with different pollution levels. Through the introduction of machine learning models, the data weight assignment is optimized, so that the system can respond to changes in air quality more intelligently.

[0032] 2. The present invention not only improves the response efficiency and control accuracy of the ventilation system, ensuring that pollutants can be discharged in a timely and effective manner, but also effectively avoids the limitations of traditional fixed ventilation systems. At the same time, the operation strategy of dynamically adjusting the air outlet not only improves the air quality, but also reduces energy consumption, achieving the dual effects of energy saving and environmental optimization. Through intelligent monitoring and control, the present invention improves the indoor air quality while ensuring the health and safety of the laboratory environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0034] Figure 1 The figure is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0036] For examples, see Figure 1 As shown, the gas flow organization intelligent control method based on air pollution detection described in this embodiment includes the following steps:

[0037] S1: Divide the laboratory into several monitoring areas, and arrange W air quality sensors in each monitoring area to obtain indoor pollutant concentration data in real time in different time periods;

[0038] S2: preprocessing the acquired indoor pollutant concentration data, and extracting the pollutant concentration diffusion rate abnormality characteristics and pollutant concentration peak fluctuation characteristics of the preprocessed indoor pollutant concentration data;

[0039] S3: According to the extracted abnormal characteristics of pollutant concentration diffusion rate and peak fluctuation characteristics of pollutant concentration, the weight assignment of indoor pollutant concentration data in each time period of the monitoring area is determined, and the comprehensive pollution index of each monitoring area is calculated after weighted average calculation;

[0040] S4: Compare and analyze the comprehensive pollution index of the monitoring area with the gradient standard threshold, divide the monitoring area into different pollution levels according to the analysis results, and determine the opening number and air volume output strategy of the corresponding air outlets in each monitoring area according to the division results;

[0041] S5: Further analyze the changes in pollution levels in each monitoring area within a fixed time period, and dynamically adjust the opening time and air volume output of the air vents based on the analysis results.

[0042] In S1, the laboratory is divided into several monitoring areas, and W air quality sensors are arranged in each monitoring area to obtain indoor pollutant concentration data in real time in different time periods, mainly including:

[0043] Functional division: According to the functional division of the laboratory, such as experimental operation area, storage area, passage area, rest area, etc. The pollutant characteristics and ventilation requirements of different functional areas may be different, so they need to be monitored separately.

[0044] Space division: Based on the physical space layout of the laboratory, the entire space is divided into several relatively independent areas, such as rooms, floors, or areas around large equipment.

[0045] Air flow path division: According to the existing air flow path, the laboratory is divided into different air flow units to better analyze the relationship between air quality and air flow.

[0046] Experimental operation area: divided into multiple monitoring areas, such as chemical experimental area, biological experimental area, physical experimental area, etc. The type and concentration of air pollutants in each area may be different, so they need to be monitored independently.

[0047] Equipment-intensive areas: The areas around large equipment or high pollution sources (such as fume hoods and chemical storage cabinets) are divided into independent monitoring areas. These areas may have higher pollutant concentrations and require intensive monitoring.

[0048] Passageways and public areas: These are divided into relatively large monitoring areas because the air flow in these areas is relatively smooth, but the spread of pollutants between different areas still needs to be monitored.

[0049] Low-Flow Areas: Zone off areas with less air flow, such as corners or enclosed areas, as these areas can become hotspots for contaminant accumulation.

[0050] By dividing the laboratory into multiple monitoring areas, pollutant sources and high-pollution areas can be more accurately located, so that targeted measures can be taken. According to the monitoring data of each area, the ventilation strategy (such as air vent switches and air volume adjustment) can be dynamically adjusted to improve the efficiency and energy efficiency of the ventilation system. Special areas (such as chemical storage areas and experimental operation areas) may produce higher concentrations of harmful gases. By independently monitoring these areas, air pollution problems can be discovered and dealt with in a timely manner to ensure the safety of laboratory personnel. Multi-area monitoring provides a richer data set that can be used for air quality trend analysis, pollution source analysis, and long-term system optimization.

[0051] W sensors per area: W air quality sensors are deployed in each monitoring area. The specific number depends on the size, shape, function and number of potential pollution sources of the area.

[0052] Place sensors in places where pollutants may be concentrated (such as near fume hoods and chemical storage areas) to ensure that the highest concentration of pollutants can be monitored. Place sensors in areas with dense human activity to ensure that the air quality in the breathing area of ​​​​people is detected. Place sensors near the inlet and outlet of air flow to monitor the effect of fresh air entering and pollutants being discharged.

[0053] Select appropriate sensors based on the types of pollutants that may be present, such as volatile organic compounds (VOCs) sensors, particulate matter (PM2.5 / PM10) sensors, CO2 sensors, etc.

[0054] Sensors collect air quality data in real time and transmit the data to the central control system via wireless or wired networks. The central control system processes and analyzes the data in real time, extracts key features and evaluates the air quality of each area. Based on the analysis results, the ventilation system is automatically adjusted to optimize air flow and pollutant emissions in each area.

[0055] S2: Preprocess the acquired indoor pollutant concentration data, and extract the pollutant concentration diffusion rate abnormality characteristics and pollutant concentration peak fluctuation characteristics of the preprocessed indoor pollutant concentration data.

[0056] Data cleaning: Delete records with a large number of missing values. This is applicable to situations where a small amount of data is missing and has little impact on the overall data. Use interpolation or statistical methods (such as mean filling, front and back value filling) to fill missing data. This is applicable to time series data or continuous monitoring data. Identify and process abnormal data points that are significantly deviated from the normal range (such as extreme values ​​caused by sensor failure). You can use statistical methods (such as z-score, box plot method) or threshold filtering based on business rules. Use smoothing filters (such as moving average method, Kalman filter) or wavelet transform methods to remove high-frequency noise in the data and retain effective signal components.

[0057] Data standardization: Convert pollutant concentration data from different sources or types to a unified scale for easy comparison and analysis. Linearly transform the data to the [0, 1] interval.

[0058] Time series processing: Smoothing time series data (such as moving average method) to reduce short-term fluctuations and highlight long-term trends. Unifying the time sampling frequency of data (such as adjusting data with different time intervals to a consistent time interval) for subsequent analysis. If you need to focus on the rate of change of data or remove trends, you can perform time series difference processing.

[0059] The pollutant concentration diffusion rate anomaly characteristics of the pre-processed indoor pollutant concentration data are extracted and the pollutant concentration diffusion rate anomaly index is generated. The method for obtaining the pollutant concentration diffusion rate anomaly index is as follows:

[0060] Obtain pollutant concentration data at time t in real time and mark it as , and the pollutant diffusion rate at time t , the calculation expression is; ; and establish the diffusion wind speed data set at time t ; The acquired pollutant concentration data, pollutant diffusion rate and diffusion wind speed data constitute the nodes in the Bayesian network, and the structure diagram of the Bayesian network is constructed, which is expressed as: ; Build a conditional probability table for each node, for example, the diffusion rate Conditional probability table , describes the distribution of pollutant diffusion rate under given concentration and environmental conditions at the previous moment, and uses historical data to estimate the parameters in the conditional probability table. The parameter estimation expression is:

[0061] ;

[0062] In the formula, In historical data The number of simultaneous occurrences, express Number of simultaneous occurrences;

[0063] For the current time t, use the Bayesian network to calculate the predicted diffusion rate distribution ; Given the actual observed diffusion rate , calculate its probability in the predicted distribution ; Calculate the pollutant concentration diffusion rate anomaly index, the expression is: ; In the formula, It is the pollutant concentration diffusion rate anomaly index.

[0064] The larger the pollutant concentration diffusion rate anomaly index is, the more serious the pollution level in the monitoring area is. This means that the rate of change of pollutant concentration in the area is far beyond the normal range, and there may be sudden pollution sources or a sharp deterioration in air quality. A significant increase in the anomaly index indicates that the current air pollution situation has deviated from the normal trend, and ventilation or other pollution control measures need to be taken in time to prevent further deterioration and harm to health.

[0065] The pollutant concentration peak fluctuation characteristics of the pre-processed indoor pollutant concentration data are extracted and the pollutant concentration peak fluctuation index is generated. The method for obtaining the pollutant concentration peak fluctuation index is:

[0066] Obtain the pollutant concentration data C(s) within the time period M, where s represents the time point in the time series. Apply discrete Fourier transform to the pollutant concentration data C(s) in the time domain and convert it into frequency domain representation C(f). The expression is:

[0067] ;

[0068] In the formula, is the pollutant concentration data spectrum value at frequency f, N is the total number of signal sample points, is the kernel function of Fourier transform, j is the imaginary unit, and the amplitude of each frequency component in the frequency domain is calculated and marked as , the expression is: ;in, is the real part of C(f), is the imaginary part of C(f), in the calculated spectrum In the process, the main frequency component is identified, that is, the frequency point whose amplitude is greater than the preset standard amplitude. , calculate its energy , the expression is ; k is the number of the main frequency, and the energy of all main frequency components is weighted and summed to generate the pollutant concentration peak fluctuation index, which is expressed as:

[0069] ;

[0070] The weight of each frequency component can be set according to its importance or the corresponding pollutant fluctuation characteristics. It is the peak fluctuation index of pollutant concentration.

[0071] The larger the peak fluctuation index of pollutant concentration, the more unstable and severe the pollution level in the monitoring area. This indicates that the pollutant concentration in the area changes frequently and with a large amplitude, and there may be sudden pollution events or periodic emission sources, resulting in significant fluctuations in air quality. A high peak fluctuation index usually means that the pollutant concentration is high and low at times, which may have a greater adverse impact on the environment and human health, so it needs to be taken seriously and control measures should be taken in a timely manner.

[0072] S3: According to the extracted abnormal characteristics of pollutant concentration diffusion rate and peak fluctuation characteristics of pollutant concentration, the weight assignment of indoor pollutant concentration data in each time period of the monitoring area is determined, and the comprehensive pollution index of each monitoring area is calculated after weighted average calculation.

[0073] The pollutant concentration diffusion rate anomaly index and the pollutant concentration peak fluctuation index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the weighted labels of the indoor pollutant concentration data in each time period in the monitoring area as the prediction target, and takes minimizing the sum of the prediction errors of the weighted labels of the indoor pollutant concentration data in all time periods as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The weighted assignment of the indoor pollutant concentration data in each time period in the monitoring area is determined according to the model output results, and the weighted average calculation is performed on them to calculate the comprehensive pollution index of each monitoring area, wherein the machine learning model is a polynomial regression model.

[0074] The method for obtaining the weight assignment of indoor pollutant concentration data in each time period of the monitoring area is as follows: from the first eigenvector training data of the trained machine learning model, the corresponding function expression is obtained: ; In the formula, is the output function of the model, is the pollutant concentration diffusion rate anomaly index, is the peak fluctuation index of pollutant concentration, Assign weights to the indoor pollutant concentration data in each time period in the monitoring area.

[0075] S4: Compare and analyze the comprehensive pollution index of the monitoring area with the gradient standard threshold, divide the monitoring area into different pollution levels according to the analysis results, and determine the opening number and air volume output strategy of the corresponding air vents in each monitoring area according to the division results.

[0076] The obtained comprehensive pollution index of each monitoring area is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the comprehensive pollution index of each monitoring area is compared with the first standard threshold and the second standard threshold respectively;

[0077] If the comprehensive pollution index of each monitoring area is greater than the second standard threshold, it means that the pollution level in the monitoring area is high, which may pose a serious threat to health and the environment. Strong ventilation measures must be taken immediately and the monitoring area shall be classified as a high pollution level;

[0078] If the comprehensive pollution index of each monitoring area is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the pollution level in the monitoring area is average and medium-intensity ventilation measures may be required to maintain air quality, and the monitoring area is classified as medium pollution level;

[0079] If the comprehensive pollution index of each monitoring area is less than the first standard threshold, it means that the pollution level in the monitoring area is low, the air quality is good, and the ventilation demand is low. It is possible to maintain a low air volume or close some air vents, and the monitoring area can be classified as a low pollution level.

[0080] Highly polluted areas: maximize the number of air outlets to increase ventilation. Set to high air volume output to ensure rapid dilution and removal of pollutants.

[0081] Medium pollution area: Open the air vents moderately to increase ventilation appropriately. Set the air volume output to a moderate level to maintain air quality within an acceptable range.

[0082] Low pollution area: Minimize the number of open air vents or close some of them. Set to low air volume output or maintain the current state to save energy and maintain a comfortable indoor environment.

[0083] According to the real-time comprehensive pollution index, the pollution level and corresponding ventilation strategy of each monitoring area are dynamically adjusted. Through continuous monitoring and feedback, the number of air outlets opened and the air volume output are optimized to ensure that each area is always at an appropriate air quality level.

[0084] S5: Further analyze the changes in pollution levels in each monitoring area within a fixed time period, and dynamically adjust the opening time and air volume output of the air vents based on the analysis results.

[0085] According to the number of air vents opened and the air volume output strategy of each monitoring area, the ventilation system can generate a preliminary airflow path to discharge pollutants, and further analyze the changes in the pollution level of each monitoring area in the subsequent fixed time period. The comprehensive pollution index in each monitoring area that is greater than or equal to the first standard threshold and less than or equal to the second standard threshold is collected, a data set is established, and the mean and standard deviation of the data set are calculated. After analyzing the data, the time to open the air vents and the air volume output are dynamically adjusted according to the analysis results.

[0086] If the mean value of the comprehensive pollution index in the data set is greater than or equal to the reference threshold value of the comprehensive pollution index mean value, and the standard deviation of the comprehensive pollution index is less than the reference threshold value of the comprehensive pollution index standard deviation, the overall comprehensive pollution index is high but the fluctuation is small, indicating that the pollution source is stable but the pollution level is serious. The opening time of the air outlet is extended and the air volume output is increased to quickly reduce the pollutant concentration;

[0087] If the mean value of the comprehensive pollution index is greater than or equal to the reference threshold value of the mean value of the comprehensive pollution index, and the standard deviation of the comprehensive pollution index is greater than or equal to the reference threshold value of the standard deviation of the comprehensive pollution index, the comprehensive pollution index is high and fluctuates greatly, indicating that there is an unstable and strong pollution source. Further extend the opening time of the air outlet, maximize the air volume output, and closely monitor the pollution changes;

[0088] If the mean of the comprehensive pollution index is less than the reference threshold of the mean of the comprehensive pollution index, and the standard deviation of the comprehensive pollution index is greater than or equal to the reference threshold of the standard deviation of the comprehensive pollution index, the comprehensive pollution index is low but fluctuates greatly, indicating that the pollution source is unstable but the average concentration is low. Keep the current air outlet opening time and maintain the existing air volume to avoid over-ventilation, while continuing to monitor fluctuations;

[0089] If the mean of the comprehensive pollution index is less than the reference threshold of the mean of the comprehensive pollution index, and the standard deviation of the comprehensive pollution index is less than the reference threshold of the standard deviation of the comprehensive pollution index, the comprehensive pollution index is low and fluctuates little, the air quality is good and stable, shorten the air vent opening time, and reduce the air volume output to save energy while maintaining low-intensity monitoring.

[0090] In this embodiment, by dividing the laboratory into multiple monitoring areas, air quality sensors are arranged to obtain pollutant concentration data in real time, and the data is preprocessed to extract the abnormal characteristics of diffusion rate and peak fluctuation characteristics. Subsequently, weighted calculations are performed based on these characteristics to generate a comprehensive pollution index for each area. After comparing and analyzing the comprehensive pollution index with the set gradient standard threshold, the area is divided into different pollution levels, and the number of air vents opened and the air volume output strategy are adjusted accordingly. Finally, the system dynamically adjusts the ventilation parameters according to the pollution changes within a fixed time period to optimize the air quality. By optimizing the air vent opening and air volume output strategy, unnecessary energy waste is avoided. Improve indoor air quality, reduce the impact of pollution on health, and ensure the safety of the laboratory environment.

[0091] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0092] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0093] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0094] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0095] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0096] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0097] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0098] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. An intelligent control method for gas flow organization based on air pollution detection, characterized in that: The following steps are involved: S1: Divide the laboratory into several monitoring areas, and arrange W air quality sensors in each monitoring area to obtain indoor pollutant concentration data in real time in different time periods; S2: preprocessing the acquired indoor pollutant concentration data, and extracting the pollutant concentration diffusion rate abnormality characteristics and pollutant concentration peak fluctuation characteristics of the preprocessed indoor pollutant concentration data; The pollutant concentration diffusion rate anomaly characteristics of the pre-processed indoor pollutant concentration data are extracted and the pollutant concentration diffusion rate anomaly index is generated. The method for obtaining the pollutant concentration diffusion rate anomaly index is as follows: Obtain pollutant concentration data at time t in real time and mark it as , and the pollutant diffusion rate at time t , the calculation expression is; ; and establish the diffusion wind speed data set at time t ; The acquired pollutant concentration data, pollutant diffusion rate and diffusion wind speed data constitute the nodes in the Bayesian network, and the structure diagram of the Bayesian network is constructed, which is expressed as: ; Construct the conditional probability table for each node, the diffusion rate Conditional probability table , using historical data, the parameters in the conditional probability table are estimated, and the parameter estimation expression is: ; In the formula, In historical data The number of simultaneous occurrences, express The number of simultaneous occurrences; for the current time t, the predicted diffusion rate distribution is calculated using the Bayesian network ; Given the actual observed diffusion rate , calculate its probability in the predicted distribution ; Calculate the pollutant concentration diffusion rate anomaly index, the expression is: ; In the formula, is the pollutant concentration diffusion rate anomaly index; The pollutant concentration peak fluctuation characteristics of the pre-processed indoor pollutant concentration data are extracted and the pollutant concentration peak fluctuation index is generated. The method for obtaining the pollutant concentration peak fluctuation index is: Obtain the pollutant concentration data C(s) within the time period M, where s represents the time point in the time series. Apply discrete Fourier transform to the pollutant concentration data C(s) in the time domain and convert it into frequency domain representation C(f). The expression is: ; In the formula, is the pollutant concentration data spectrum value at frequency f, N is the total number of signal sample points, is the kernel function of Fourier transform, j is the imaginary unit, and the amplitude of each frequency component in the frequency domain is calculated and marked as , the expression is: ;in, is the real part of C(f), is the imaginary part of C(f), in the calculated spectrum In the process, the main frequency component is identified, that is, the frequency point whose amplitude is greater than the preset standard amplitude. , calculate its energy , the expression is ; k is the number of the main frequency, and the energy of all main frequency components is weighted and summed to generate the pollutant concentration peak fluctuation index, which is expressed as: ; is the weight of each frequency component, is the peak fluctuation index of pollutant concentration; S3: According to the extracted abnormal characteristics of pollutant concentration diffusion rate and peak fluctuation characteristics of pollutant concentration, the weight assignment of indoor pollutant concentration data in each time period of the monitoring area is determined, and the comprehensive pollution index of each monitoring area is calculated after weighted average calculation; S4: Compare and analyze the comprehensive pollution index of the monitoring area with the gradient standard threshold, divide the monitoring area into different pollution levels according to the analysis results, and determine the opening number and air volume output strategy of the corresponding air outlets in each monitoring area according to the division results; S5: Further analyze the changes in pollution levels in each monitoring area within a fixed time period, and dynamically adjust the opening time and air volume output of the air vents based on the analysis results.

2. The method for intelligent control of gas flow organization based on air pollution detection according to claim 1 is characterized in that: In S3, the pollutant concentration diffusion rate anomaly index and the pollutant concentration peak fluctuation index are converted into the first eigenvector, and the first eigenvector is used as the input of the machine learning model. The machine learning model uses each group of first eigenvectors to predict the weighted labels of the indoor pollutant concentration data in each time period of the monitoring area as the prediction target, and takes minimizing the sum of the prediction errors of the weighted labels of the indoor pollutant concentration data in all time periods as the training target. The machine learning model is trained until the sum of the prediction errors converges and the model training is stopped. The weighted assignment of the indoor pollutant concentration data in each time period of the monitoring area is determined according to the model output results, and the weighted average calculation is performed on them to calculate the comprehensive pollution index of each monitoring area, wherein the machine learning model is a polynomial regression model.

3. The method for intelligent control of gas flow organization based on air pollution detection according to claim 2 is characterized in that: In S4, the comprehensive pollution index of the monitoring area is compared and analyzed with the gradient standard threshold, and the monitoring area is divided into different pollution degree levels according to the analysis results; The obtained comprehensive pollution index of each monitoring area is compared with the gradient standard threshold, the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and the comprehensive pollution index of each monitoring area is compared with the first standard threshold and the second standard threshold respectively; If the comprehensive pollution index of each monitoring area is greater than the second standard threshold, the monitoring area will be classified as a high pollution level; if the comprehensive pollution index of each monitoring area is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the monitoring area will be classified as a medium pollution level; if the comprehensive pollution index of each monitoring area is less than the first standard threshold, the monitoring area will be classified as a low pollution level.

4. The method for intelligent control of gas flow organization based on air pollution detection according to claim 1 is characterized in that: In S5, the pollution level change of each monitoring area within a fixed time period is further analyzed, and the opening time and air volume output of the air vents are dynamically adjusted according to the analysis results; According to the number of air vents opened and the air volume output strategy of each monitoring area, the ventilation system can generate a preliminary airflow path to discharge pollutants, and further analyze the changes in the pollution level of each monitoring area in the subsequent fixed time period. The comprehensive pollution index in each monitoring area that is greater than or equal to the first standard threshold and less than or equal to the second standard threshold is collected, a data set is established, and the mean and standard deviation of the data set are calculated. After analyzing the data, the time to open the air vents and the air volume output are dynamically adjusted according to the analysis results.

5. The method for intelligent control of gas flow organization based on air pollution detection according to claim 4 is characterized in that: If the mean value of the comprehensive pollution index in the data set is greater than or equal to the reference threshold value of the comprehensive pollution index mean value, and the standard deviation of the comprehensive pollution index is less than the reference threshold value of the comprehensive pollution index standard deviation, it indicates that the pollution source is stable but the pollution level is serious. The opening time of the air outlet is extended and the air volume output is increased to quickly reduce the pollutant concentration; If the mean value of the comprehensive pollution index is greater than or equal to the reference threshold value of the mean value of the comprehensive pollution index, and the standard deviation of the comprehensive pollution index is greater than or equal to the reference threshold value of the standard deviation of the comprehensive pollution index, it indicates that there is an unstable and strong pollution source. Further extend the opening time of the air outlet, maximize the air volume output, and closely monitor the pollution changes; If the mean value of the comprehensive pollution index is less than the reference threshold value of the mean value of the comprehensive pollution index, and the standard deviation of the comprehensive pollution index is greater than or equal to the reference threshold value of the standard deviation of the comprehensive pollution index, it indicates that the pollution source is unstable but the average concentration is low. Keep the current air outlet opening time and the existing air volume, and continue to monitor the fluctuation; If the mean of the comprehensive pollution index is less than the reference threshold of the mean of the comprehensive pollution index, and the standard deviation of the comprehensive pollution index is less than the reference threshold of the standard deviation of the comprehensive pollution index, the air quality is good and stable, shorten the air vent opening time, and reduce the air volume output.

Images