Laboratory air conditioning system and control method thereof

Abstract

The invention discloses a laboratory air conditioning system and a control method thereof, and particularly relates to the technical field of air conditioning. Intelligent monitoring and dynamic adjustment of air quality in an experiment area are achieved through coordinated work of the central control module, the zone control module and the zone sensor module, the system evaluates the air pollution severity degree of each area in real time according to air quality sensing data, and through dynamic linkage of the air purification module, the air flow control module and the temperature and humidity adjusting module, the air quality of the experiment area can be monitored and adjusted dynamically. In addition, by predicting future air quality changes and adjusting air quality parameters and optimizing energy consumption in advance according to prediction results, it is ensured that energy consumption is reduced while the air quality meets experimental requirements, and the precision and efficiency of laboratory air quality management are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioning, and in particular to a laboratory air conditioning system and a control method thereof. Background Art

[0002] Laboratory air conditioning systems and control methods are widely used in scientific research, medicine, chemical industry and other fields to ensure that the laboratory's air quality, temperature and humidity, air flow speed and cleanliness meet specific experimental requirements. With the increasing complexity of experimental content, especially in interdisciplinary experiments or multi-functional laboratories, the air quality requirements of different experiments vary greatly, and traditional air conditioning systems are insufficient in fine control and response to sudden pollutant diffusion. Multi-functional laboratories often accommodate a variety of experimental activities such as chemistry, biology, and physics, and each experiment has significant differences in the requirements for air quality, temperature and humidity, and cleanliness. Some experiments may release volatile organic compounds (VOCs), particulate matter, or bioaerosols. If these pollutants are not isolated or treated in time, they are likely to affect the accuracy of other experiments and even threaten the health of experimenters. Traditional air conditioning systems cannot respond quickly when sudden pollutants appear, cannot accurately isolate pollution sources by region, and lack the ability to dynamically monitor multiple airflow changes in complex experimental environments. In addition, existing systems often face problems of insufficient adjustment accuracy and delayed response under high load, resulting in inaccurate airflow speed and direction control, which makes it impossible to prevent pollutant diffusion in time. This limitation of being unable to achieve real-time dynamic regulation is particularly prominent in multi-functional laboratories, and has become a major hidden danger that affects the reliability of experimental results and the safety of experimenters. Summary of the invention

[0003] The object of the present invention is to provide a laboratory air conditioning system and a control method thereof to solve the deficiencies in the background technology.

[0004] In order to achieve the above object, the present invention provides the following technical solution: a laboratory air conditioning system, including a central control module, a regional sensor module, a partition control module, an air purification module and an airflow control module: The central control module is used to control the entire system and communicate with each partition control module to achieve linkage control of air quality parameters in multiple areas; Regional sensor modules, including temperature and humidity sensors, air quality sensors, and airflow velocity sensors, are distributed in various experimental areas of the laboratory to detect and feedback the air quality, temperature and humidity, and airflow data of each area in real time; The partition control module is located in each experimental area of ​​the laboratory. It is used to determine the severity of air pollution in each experimental area based on the data fed back by the regional sensor module, and dynamically adjust the air purification module, temperature and humidity adjustment module and airflow control module of the experimental area according to the severity of air pollution to achieve independent air quality control; An air purification module, including a high-efficiency air filter, an activated carbon filter and a UV disinfection device, is used to remove volatile organic compounds, bioaerosols and particulate matter from the air; The airflow control module is used to adjust the airflow speed and direction of each area according to the changes in the severity of air pollution in each experimental area within a fixed time period, ensuring the orderly flow of air in each area and avoiding the spread of pollutants.

[0005] Preferably, in the partition control module, the pollutant concentration fluctuation index is generated after analyzing the abnormal fluctuation of the time change rate of the pollutant concentration in each experimental area. The method for obtaining the pollutant concentration fluctuation index is: Set a time window x, that is, calculate the number of data points contained in the volatility index, assign a weight wi to each data point in the given time window, and mark the pollutant concentration at x time points as ;in, is the earliest data point, For the latest data point, calculate the weighted moving average EF of the pollutant concentration, the expression is: ; In the formula, is the pollutant concentration at the i-th time point, and the pollutant concentration fluctuation index is defined as the latest actual concentration value The difference with EF is expressed as: ; In the formula, is the pollutant concentration fluctuation index.

[0006] Preferably, in the partition control module, the air flow pattern in the area is monitored using the air flow velocity sensor data, and the air flow diffusion interference index is generated according to the interference degree of the air flow velocity and direction changes. The method for obtaining the air flow diffusion interference index is: Obtain the time series data of airflow velocity and direction from the airflow velocity sensor, set the sampling frequency f, collect the data for the past T seconds, and select a time series representing the steady state of airflow from the historical data as the baseline airflow pattern, recorded as ;in Represents the speed and direction data at the nth moment in the reference airflow pattern, and records the sequence data of the airflow speed and direction in the current time period, recorded as ,in represents the speed and direction data at the mth moment in the current airflow pattern. Assuming that each data point contains the airflow speed v and the airflow direction θ, then the i-th moment reference point With the current airflow point The Euclidean distance between for: ; Construct an n×m DTW matrix D, where D(i,j) represents the reference airflow pattern With the current airflow model The cumulative minimum matching cost is initialized to D(0,0)=0. For each i∈[1,n] and j∈[1,m], the recursive formula is used to calculate each element in the DTW matrix: ; The airflow diffusion interference index is defined as the final matching cost between the reference airflow pattern and the current airflow pattern, and the expression is: ; Where SD is the airflow diffusion interference index, D(n,m) is the final element of the matrix D, and D(n,m) represents the minimum matching cost of the two airflow patterns.

[0007] Preferably, the pollutant concentration fluctuation index and the airflow diffusion interference 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 air pollution severity value label in each experimental area as the prediction target, and takes minimizing the sum of prediction errors of the air pollution severity value labels in all experimental areas as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The air pollution severity value in each experimental area is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0008] Preferably, the acquired air pollution severity value in each experimental area is compared with a 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 air pollution severity value in each experimental area is compared with the first standard threshold and the second standard threshold respectively; If the air pollution severity value in the experimental area is greater than the second standard threshold, it means that the air pollution severity in the experimental area is high. At this time, a high pollution degree signal is generated, and the experimental area is divided into a high pollution area; If the air pollution severity value in the experimental 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 air pollution severity in the experimental area is average. At this time, a medium pollution degree signal is generated and the experimental area is divided into a medium pollution area; If the air pollution severity value in the experimental area is less than the first standard threshold, it means that the air pollution severity in the experimental area is low. At this time, a low pollution degree signal is generated and the experimental area is divided into a low pollution area.

[0009] Preferably, in the airflow control module, the airflow velocity v is set to control the diffusion trend of pollutants. The airflow velocity should be proportional to the rate of change of the severity of air pollution S. When the severity of pollution increases, the airflow velocity is increased to quickly extract pollutants; when the severity of pollution is low or stable, the airflow velocity is maintained at a low level to save energy. The specific formula is: ; is the airflow velocity at the current time t, is the basic airflow velocity used to maintain air flow and prevent pollutant accumulation. is the airflow velocity adjustment factor, which is used to control the sensitivity of the impact of changes in pollution severity on airflow velocity. is the rate of change of air pollution severity S; The airflow angle θ is set to deviate from the pollution source area. The airflow direction adjustment formula is: ; The airflow direction angle at the current time t, is the angle of the target exhaust direction, is the airflow direction adjustment coefficient, which is used to control the adjustment range of airflow direction due to pollution severity. is the change in the severity of air pollution.

[0010] Preferably, in the airflow control module, when >0 and ΔS>0, it means that the pollution is aggravated and the air flow speed should be increased rapidly And adjust the airflow direction Close to the exhaust system to avoid the spread of pollutants; <0 and ΔS<0, indicating that pollution is alleviated, and the air flow speed is gradually reduced. To base speed And restore the original direction to maintain the stability of regional air flow.

[0011] The present invention also provides a laboratory air conditioning control method, comprising: S1: Monitor the air quality, temperature, humidity and airflow data of the experimental area through the regional sensor module, and feed the data back to the central control module; S2: Analyze the severity of air pollution in each experimental area based on the detection data, and dynamically adjust the air purification module, temperature and humidity adjustment module and airflow control module in each area; S3: When it is detected that the severity of air pollution in the experimental area changes abnormally, the airflow speed and direction of each experimental area are adjusted in real time; S4: Predict the air quality change trend in each experimental area, and adjust the air quality parameters in advance according to the prediction results to optimize energy consumption.

[0012] Preferably, in S4, the air quality data of each experimental area is collected, the missing values ​​and outliers in the data are processed, and feature inputs are generated. The feature data are then standardized and divided into a training set and a test set in chronological order. The radial basis kernel function of the support vector regression model is selected to process nonlinear data. The model parameters are optimized through grid search and cross-validation to maintain the best prediction effect under different settings. After the model is trained using the training set, the latest air quality data is input to predict the trend of air quality changes in the future. If the prediction shows that the air quality has deteriorated, the purification intensity is increased in advance, and the airflow direction and temperature and humidity are adjusted; if the prediction remains good, the purification and adjustment frequency is reduced to optimize energy consumption.

[0013] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. Through the coordinated work of the central control module, the partition control module and the regional sensor module, the system can monitor the concentration of pollutants, temperature and humidity, and airflow parameters in the air in real time, and respond quickly when the severity of air pollution changes. When the air purification module and the airflow control module detect changes in the flow pattern of air pollutants, they can independently adjust the purification intensity, airflow speed and direction by region to ensure the cleanliness and safety of the air in the laboratory, avoid the spread of pollutants in the multifunctional experimental environment, and thus improve the reliability of the experimental results.

[0014] 2. The present invention predicts the future trend of air quality by introducing trend prediction and machine learning algorithms, using the support vector regression model, and adjusts the parameters of air purification, temperature and humidity, and airflow control in advance according to the prediction results to optimize the system energy consumption. The system dynamically divides the pollution level of the experimental area and automatically adjusts the air flow pattern according to the pollutant concentration fluctuation index and the airflow diffusion interference index, thereby reducing energy consumption while meeting air quality requirements, which not only ensures the safety and cleanliness of the multifunctional experimental environment, but also significantly improves energy efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] 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.

[0016] Figure 1 The figure is a flow chart of the method of the present invention.

[0017] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0018] 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.

[0019] Example 1, please refer to Figure 1 and Figure 2 As shown, a laboratory air conditioning system described in this embodiment includes a central control module, a regional sensor module, a partition control module, an air purification module and an airflow control module: The central control module is used to control the entire system and communicate with each partition control module to achieve linkage control of air quality parameters in multiple areas; Regional sensor modules, including temperature and humidity sensors, air quality sensors, and airflow velocity sensors, are distributed in various experimental areas of the laboratory to detect and feedback the air quality, temperature and humidity, and airflow data of each area in real time; The partition control module is located in each experimental area of ​​the laboratory. It is used to determine the severity of air pollution in each experimental area based on the data fed back by the regional sensor module, and dynamically adjust the air purification module, temperature and humidity adjustment module and airflow control module of the experimental area according to the severity of air pollution to achieve independent air quality control; An air purification module, including a high-efficiency air filter, an activated carbon filter and a UV disinfection device, is used to remove volatile organic compounds, bioaerosols and particulate matter from the air; The airflow control module is used to adjust the airflow speed and direction of each area according to the changes in the severity of air pollution in each experimental area within a fixed time period, ensuring the orderly flow of air in each area and avoiding the spread of pollutants.

[0020] The central control module is connected to the sensor modules in each partition to continuously receive air quality data from each area of ​​the laboratory, including temperature and humidity, air flow velocity, concentrations of harmful gases (such as VOCs), particulate matter and bioaerosols, etc. The layout of these sensors covers the key areas of the laboratory, enabling the central control module to fully and real-time grasp the air quality status of each area.

[0021] The central control module has a built-in data processing unit for analyzing and integrating monitoring data from different areas. The unit uses preset standards and data models to analyze and judge temperature, humidity, air cleanliness, and airflow conditions, and identifies any air quality anomalies that deviate from the set standards. Through intelligent algorithms based on historical data and real-time information, the data processing unit can also predict air quality trends and adjust the air conditioning system in advance.

[0022] When the data processing unit detects that the air quality parameters of a certain experimental area exceed the set threshold (for example, the concentration of VOCs or particulate matter is too high), the command sending unit of the central control module immediately sends a command to the partition control module of the area. After receiving the command, the partition control module executes specific air conditioning measures, including starting the air purification device, adjusting the airflow direction, adjusting the temperature and humidity, etc., so as to quickly improve the air quality. This process is carried out through a feedback loop of real-time data monitoring to ensure the rapid response and accuracy of the control measures.

[0023] The central control module is equipped with a linkage response module. When there is a linkage demand or emergency situation (such as harmful gas leakage, experimental cross-contamination risk) in the air quality parameters of multiple regions, it coordinates the control modules of each partition to perform system-level regulation. For example, when VOCs leaks in the chemical experiment area and the concentration exceeds the standard, the central control module can automatically switch the area to negative pressure mode, while increasing positive pressure in other sensitive areas (such as biological experiment areas) to ensure that pollutants do not spread to other areas. In addition, the linkage response module can increase the power of the air purification equipment in the area and adjust the airflow direction of adjacent areas to form an effective isolation zone.

[0024] By continuously accumulating and analyzing historical data on air quality, the central control module can gradually learn the air control needs and change patterns of each experimental area. With the help of intelligent prediction algorithms, the central control module can identify possible air quality fluctuations in advance and make preventive adjustments. For example, the purification efficiency of the area can be appropriately improved before the peak period of chemical experiments, or advance control can be performed during periods when temperature and humidity may change dramatically. This can not only improve the control accuracy, but also effectively reduce system energy consumption.

[0025] The central control module also has fault detection and alarm functions. When a partition control module or sensor module fails or fails to respond, the central control module can promptly identify and issue an alarm. The system automatically activates backup plans or adjusts the airflow and air purification strategies in the surrounding areas to ensure the overall safety of the laboratory.

[0026] In the multifunctional laboratory air conditioning system, the regional sensor module is a key component to ensure accurate control of air quality. It is responsible for real-time detection and feedback of data such as temperature, humidity, air quality and airflow in each experimental area. The core task of this module is to continuously and finely collect air parameters in each experimental area through a high-precision sensor network, and transmit the data to the central control module to achieve intelligent monitoring and regulation of air quality.

[0027] The temperature and humidity sensor is an important part of the regional sensor module. It is responsible for real-time detection of the temperature and humidity levels of each experimental area to ensure the stability of the experimental environment. Changes in temperature and humidity will directly affect the results of the experiment. Especially in experiments that require constant temperature or humidity, accurate temperature and humidity control is essential. The sensor module operates in the following ways: Temperature detection: Using components such as thermistors or thermocouples to detect temperature changes in the environment and output resistance values ​​or voltage changes to obtain accurate temperature readings.

[0028] Humidity detection: Capacitive humidity sensors are usually used to detect the effect of moisture in the air on the capacitor and convert it into an electrical signal. The humidity sensor in the system can detect different indicators such as absolute humidity and relative humidity to meet the needs of various experiments.

[0029] The detected temperature and humidity data will be transmitted to the central control module in real time for the system to optimize the air conditioning strategy. For example, when the temperature and humidity exceed the set range, the system will instruct the partition control module in the corresponding area to turn on the temperature and humidity adjustment device to quickly restore the environment to an appropriate level.

[0030] Air quality sensors are important components to ensure the safety of the experimental environment. They can detect a variety of pollutants in the air, such as volatile organic compounds (VOCs), particulate matter (PM2.5, PM10), carbon dioxide, harmful gases (such as ammonia, formaldehyde), etc., to ensure that experimenters are not affected by harmful substances and avoid cross-contamination in experiments. Specific detection methods include: VOCs detection: Usually a photoionization sensor (PID) or a metal oxide sensor (MOS) is used to sensitively detect changes in organic gas concentration in the air and provide real-time feedback on the concentration.

[0031] Particle detection: Using laser scattering or optical particle sensors, particle sensors can detect PM2.5 and PM10 concentrations in the air, suitable for chemical or biological experimental environments to prevent the spread of particulate matter.

[0032] Carbon dioxide detection: Use infrared sensors (NDIR) to achieve high-precision detection to ensure that the carbon dioxide concentration in the experimental environment is maintained within a safe range, which is especially important in a closed experimental environment.

[0033] Once the air quality sensor detects that harmful substances exceed the standard, the data will be quickly fed back to the central control module, and the system will automatically instruct the air purification equipment in the area to operate, strengthen airflow control, and prevent the spread of pollutants.

[0034] The air velocity sensor is used to detect the flow speed and direction of air in the experimental area to ensure reasonable air flow distribution, thereby effectively isolating potential pollution sources and maintaining the cleanliness of the experimental area. The main working principles of the air velocity sensor include: Thermal air velocity sensor: uses a heating element to detect the effect of air flow on temperature changes to measure air velocity. This method is suitable for providing high-precision measurements in low-speed airflow environments.

[0035] Ultrasonic wind speed sensor: determines wind speed and direction by measuring the time difference of ultrasonic pulses propagating in the air. It is suitable for laboratory environments that require high sensitivity.

[0036] The air velocity sensor can help the central control module determine the air flow status of different experimental areas. If a certain area needs to maintain a negative pressure or positive pressure environment, the data of the air velocity sensor will be used as the basis for regulation. For example, in the chemical experiment area, the negative pressure environment can prevent the diffusion of volatile harmful gases, while in the biological experiment area, positive pressure is maintained to prevent external pollution from entering. The central control module adjusts the airflow direction and speed according to the airflow sensor data to ensure that the regional air flow meets the experimental requirements.

[0037] The regional sensor module forms a complete air quality monitoring network through continuous data collection. Each sensor transmits data to the central control module in real time. Through a unified feedback mechanism, the system realizes closed-loop management between data collection and control decision-making. The high frequency of data feedback ensures that the dynamic changes of parameters such as air quality, temperature and humidity, and air flow speed can be responded to in a very short time. This mechanism ensures that each experimental area can start the control equipment in time when the air quality is abnormal, so as to keep the experimental environment stable.

[0038] The zone control module is located in each experimental area of ​​the laboratory and is used to determine the severity of air pollution in each experimental area based on the data fed back by the zone sensor module. Specifically: The partition control module receives real-time data such as pollutant concentration (such as VOCs, particulate matter, etc.), temperature and humidity, and airflow speed in the experimental area from the regional sensor module. An air quality benchmark is established for each experimental area, including basic pollutant concentration, normal airflow status and other data. These benchmarks will be used to detect whether abnormal increases occur.

[0039] After analyzing the abnormal fluctuation of the time change rate of pollutant concentration in each experimental area, the pollutant concentration fluctuation index is generated. The method for obtaining the pollutant concentration fluctuation index is as follows: Set a time window x, that is, the number of data points included in the calculation of the volatility index. If x=5, it means that the current volatility index is based on the pollutant concentration at the past 5 time points. Each data point in the given time window is assigned a weight wi, so that the weight increases from the earliest to the latest. The sum of the weights is 1 for normalization. The common way to set weights is to set each weight to: ; where i represents the position of the i-th data point in the window, is a normalization constant that makes the sum of all weights equal to 1, and marks the pollutant concentration at x time points as ;in, is the earliest data point, For the latest data point, calculate the weighted moving average EF of the pollutant concentration, the expression is: ; In the formula, is the pollutant concentration at the i-th time point, and the pollutant concentration fluctuation index is defined as the latest actual concentration value The difference with EF is expressed as: ; In the formula, is the pollutant concentration fluctuation index. The larger the pollutant concentration fluctuation index, the greater the deviation between the current concentration and the weighted moving average concentration, and the more drastic the fluctuation; when the fluctuation index is small, it means that the pollutant concentration changes relatively smoothly.

[0040] The air flow pattern in the area is monitored using the air velocity sensor data, and the air flow diffusion interference index is generated according to the interference degree of the air flow velocity and direction changes. The method for obtaining the air flow diffusion interference index is: Obtain time series data of airflow velocity and direction from the airflow velocity sensor. Set the sampling frequency f (e.g., once per second) and collect data for the past T seconds. Select a time series representing the steady state of airflow from the historical data as the baseline airflow pattern, denoted as ;in Represents the speed and direction data at the nth moment in the reference airflow pattern, and records the sequence data of the airflow speed and direction in the current time period, recorded as ,in represents the speed and direction data at the mth moment in the current airflow pattern. Assuming that each data point contains the airflow speed v and the airflow direction θ, then the i-th moment reference point With the current airflow point The Euclidean distance between for: ; To avoid the periodic influence of the direction angle, the angle difference The minimum angle difference is calculated within the range of [−180°, 180°] to ensure more accurate similarity calculation. Construct an n×m DTW matrix D, where D(i,j) represents the reference airflow pattern With the current airflow model The cumulative minimum matching cost is initialized to D(0,0)=0. For each i∈[1,n] and j∈[1,m], the recursive formula is used to calculate each element in the DTW matrix: ; The formula is used to select the minimum cost path from the current point (i, j) to the previous three adjacent points, and accumulate the Euclidean distance of the path; The airflow diffusion interference index is defined as the final matching cost between the reference airflow pattern and the current airflow pattern (i.e., the lower right corner element of the DTW matrix), and the expression is: ; Where SD is the airflow diffusion interference index, D(n,m) is the final element of the matrix D, and D(n,m) represents the minimum matching cost of two airflow patterns. The higher the airflow diffusion interference index, the greater the deviation of the current airflow pattern from the steady-state pattern and the more obvious the interference; the lower the index, the more similar the current airflow is to the reference pattern and the airflow is stable.

[0041] The pollutant concentration fluctuation index and the airflow diffusion interference 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 predicts the air pollution severity value label in each experimental area with each group of first eigenvectors as the prediction target, and minimizes the sum of prediction errors of the air pollution severity value labels in all experimental areas as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The air pollution severity value in each experimental area is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0042] The method for obtaining the severity value of air pollution in each experimental 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 fluctuation index, SD is the airflow diffusion interference index, is the severity of air pollution in each experimental area.

[0043] The acquired air pollution severity value in each experimental 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 air pollution severity value in each experimental area is compared with the first standard threshold and the second standard threshold respectively; If the air pollution severity value in the experimental area is greater than the second standard threshold, it means that the air pollution severity in the experimental area is high. At this time, a high pollution degree signal is generated, and the experimental area is divided into a high pollution area; If the air pollution severity value in the experimental 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 air pollution severity in the experimental area is average. At this time, a medium pollution degree signal is generated and the experimental area is divided into a medium pollution area; If the air pollution severity value in the experimental area is less than the first standard threshold, it means that the air pollution severity in the experimental area is low. At this time, a low pollution degree signal is generated and the experimental area is divided into a low pollution area.

[0044] The dynamic adjustment of the air purification module includes: High pollution area: When the experimental area is judged to be highly polluted, the air purification module will immediately run at the highest intensity, start the high-efficiency air filter (HEPA filter) and activated carbon filter device, and increase the purification frequency to quickly reduce the concentration of pollutants. Medium pollution area: The air purification module runs at a medium intensity, and the filter device is started intermittently to maintain a low pollution level and prevent the accumulation of pollutants. Low pollution area: The air purification module runs at the lowest intensity, or only starts the filter during the preset time period to keep the air quality stable and reduce energy consumption.

[0045] The dynamic adjustment of the temperature and humidity control module includes: Temperature and humidity fluctuation response: Automatically adjust the temperature and humidity control module according to the temperature and humidity data fed back by the sensor. If the temperature and humidity in the high-pollution area are higher than the set value, the dehumidification and cooling device will be started to prevent further pollution spread; if it is lower than the set value, humidification and heating will be performed accordingly. Temperature and humidity maintenance strategy: The medium-pollution and low-pollution areas maintain the temperature and humidity in the preset range according to demand to avoid the impact of temperature and humidity changes on air quality and the diffusion rate of pollutants.

[0046] The dynamic adjustment of the airflow control module includes: Airflow isolation: In high-pollution areas, the airflow control module will prioritize negative pressure control to prevent pollutants from spreading to other areas, adjust the airflow direction and control the airflow speed to ensure that pollutants are concentrated in the area. Airflow guidance: In medium-pollution areas, the airflow speed is controlled by the airflow guide device to form a positive pressure environment to prevent external pollutants from entering. Airflow stabilization: Low-pollution areas maintain natural airflow or slight positive pressure, reduce airflow disturbances to stabilize air quality, and reduce adjustment costs.

[0047] An air purification module, including a high-efficiency air filter, an activated carbon filter and a UV disinfection device, is used to remove volatile organic compounds, bioaerosols and particulate matter from the air; High-efficiency air filters (HEPA filters) are designed to capture fine particles (such as PM2.5, PM10) and some bioaerosols in the air. HEPA filters use high-density fiber mesh and have a filtration efficiency of more than 99.97% for particles of 0.3 microns and above. When air passes through the fiber mesh, the particles are blocked and adsorbed, and the purified air is output to the laboratory. HEPA filters are particularly suitable for filtering dust, smoke and microbial particles generated during the experiment, and can operate efficiently under medium and high pollution conditions.

[0048] Activated carbon filters are used to absorb volatile organic compounds (VOCs) and some odors and chemical gases in the air. There are a large number of tiny pores on the surface of activated carbon, which can absorb gas molecules, especially organic compound molecules such as VOCs, formaldehyde, and benzene. When the air passes through the activated carbon layer, these pollutants are adsorbed in the carbon pores, thereby achieving air purification. When the VOCs concentration in the experimental area is high (such as the chemical experimental area), the activated carbon filter is activated first to reduce the impact of harmful gases on experimental results and personnel health.

[0049] Ultraviolet disinfection devices are specially used to kill microorganisms, bacteria and viruses in the air to prevent the spread of biological contamination in the experimental environment. Ultraviolet (UV-C) light waves can destroy the DNA or RNA structure of microorganisms, preventing their growth and reproduction. When air flows through the UV lamp, the microorganisms carried are irradiated by ultraviolet rays, and the killing rate can reach more than 99%. Ultraviolet disinfection devices are mainly used in biological experimental areas to ensure a sterile environment of the air and prevent microbial contamination from interfering with the experimental process.

[0050] Dynamic linkage working mode: High pollution situation: When the air pollution severity value is high, the air purification module will automatically start the high-efficiency mode of HEPA filter, activated carbon filter and ultraviolet disinfection device to ensure that all kinds of pollutants are removed quickly and effectively. Medium pollution situation: The air purification module selectively starts the medium intensity of HEPA and activated carbon filters to maintain air quality and reduce energy consumption. Low pollution situation: HEPA filter runs at low intensity, activated carbon and ultraviolet disinfection devices are on standby, and are only started briefly when slight VOCs or biological contamination is detected.

[0051] The airflow control module is used to adjust the airflow speed and direction of each area according to the changes in the severity of air pollution in each experimental area within a fixed time period, ensuring the orderly flow of air in each area and avoiding the spread of pollutants.

[0052] The airflow velocity v is set to control the diffusion trend of pollutants. The airflow velocity should be proportional to the rate of change of the severity of air pollution S. When the severity of pollution increases, the airflow velocity should be appropriately increased to quickly remove pollutants; when the severity of pollution is low or relatively stable, the airflow velocity should be maintained at a low level to save energy. The specific formula is: ; is the airflow velocity at the current time t, is the basic airflow velocity used to maintain air flow and prevent pollutant accumulation. is the airflow velocity adjustment factor, which is used to control the sensitivity of the impact of changes in pollution severity on airflow velocity. It is the rate of change of the severity of air pollution S. It is usually calculated by the difference in pollution levels within a time period t, indicating the speed at which the pollution situation is aggravated or alleviated.

[0053] The goal of adjusting the airflow direction is to prevent pollutants from spreading to other areas, especially in highly polluted areas, to ensure that the airflow is directed toward the exhaust or filtration system. Set the airflow angle θ to deviate from the pollution source area and maintain the consistency of the overall airflow direction. The airflow direction adjustment formula is: ; The airflow direction angle at the current time t, The angle of the target exhaust direction, usually pointing to the direction of the external exhaust system or filter. is the airflow direction adjustment coefficient, which is used to control the adjustment range of airflow direction due to pollution severity. It is the change in the severity of air pollution. When the severity of pollution increases, it will move towards the target direction to prevent the pollution from spreading to adjacent areas.

[0054] During system operation, the air flow speed and direction adjustment strategies are as follows: Pollution rises: When >0 and ΔS>0, it means that the pollution is getting worse, and the system should quickly increase the airflow speed And adjust the airflow direction Stay close to the exhaust system to prevent pollutants from spreading to other areas.

[0055] Pollution reduction: When <0 and ΔS<0, it means that the pollution is relieved and the air flow speed can be gradually reduced. To base speed And restore the original direction to maintain the stability of regional air flow.

[0056] In this embodiment, the central control module, as the core of the system, is responsible for communicating with each partition control module and coordinating the linkage control of air quality parameters in each area. The regional sensor modules (including temperature and humidity, air quality, and airflow velocity sensors) in each experimental area monitor the air quality data in real time and feed back to the corresponding partition control module. The partition control module analyzes the severity of air pollution based on the detection data, and dynamically adjusts the air purification module, temperature and humidity adjustment module, and airflow control module in each area. The air purification module removes volatile organic compounds (VOCs), bioaerosols, and particulate matter through high-efficiency filters, activated carbon filters, and ultraviolet disinfection devices to ensure clean air; the airflow control module adjusts the airflow speed and direction according to changes in the severity of pollution to ensure the orderliness of air flow and prevent pollutants from spreading to other areas, thereby achieving independent and precise air quality control.

[0057] Embodiment 2, a laboratory air conditioning control method described in this embodiment comprises the following steps: S1: Monitor the air quality, temperature, humidity and airflow data of the experimental area through the regional sensor module, and feed the data back to the central control module; S2: Analyze the severity of air pollution in each experimental area based on the detection data, and dynamically adjust the air purification module, temperature and humidity adjustment module and airflow control module in each area; S3: When it is detected that the severity of air pollution in the experimental area changes abnormally, the airflow speed and direction of each experimental area are adjusted in real time; S4: Predict the air quality change trend in each experimental area, and adjust the air quality parameters in advance according to the prediction results to optimize energy consumption.

[0058] Step S4 specifically includes: collecting air quality related data from the sensor modules in each experimental area, including pollutant concentration (such as VOCs, PM2.5), temperature and humidity, air flow speed, etc. It is assumed that the time interval for collecting data is fixed (for example, once per minute).

[0059] Interpolate or remove missing values ​​and filter outliers. Convert raw air quality data into feature inputs for the SVR model, such as the average, maximum, minimum, and airflow change trends of pollutant concentrations over the past 24 hours, as features that affect future air quality. Standardize feature data to make them within a uniform range of values ​​(for example, using a standard normal distribution with a mean of 0 and a standard deviation of 1) to improve the stability and accuracy of model training.

[0060] The collected data is divided into a training set and a test set in chronological order. The training set is used to train the SVR model, and the test set is used to evaluate the prediction effect of the model. For example, the first 80% of the data is used as the training set, and the last 20% of the data is used as the test set.

[0061] Select the kernel function. The kernel function of the SVR model (such as linear kernel, radial basis kernel RBF, polynomial kernel) will affect the nonlinear fitting ability of the model. For air quality prediction, the radial basis function (RBF) kernel can be selected because it is suitable for processing nonlinear data. Use grid search and cross-validation to optimize the key parameters of the SVR model: Penalty coefficient CCC: controls the model's tolerance to training data errors. The higher the CCC, the more the model tends to reduce errors, but may lead to overfitting. Kernel function parameters (such as γ of the RBF kernel): control the nonlinear fitting ability of the model. The larger the γ, the stronger the nonlinearity of the model. Perform cross-validation in the training set to ensure that the model can maintain good prediction results under different parameter combinations.

[0062] The optimized SVR model is trained using the training set data, and the model learns the historical trends and change patterns of air quality parameters (such as pollutant concentrations). After training, the model is evaluated using the test set to observe the prediction error (such as root mean square error RMSE or mean absolute error MAE) to ensure that the model can accurately predict air quality changes.

[0063] In actual operation, the latest air quality data (such as pollutant concentration, temperature and humidity, air flow speed, etc.) of each experimental area is input, and the model will output the predicted air quality value for a certain period of time in the future (such as the next hour). Multi-step predictions can be made for air quality at different time points (such as pollutant concentrations in the next 1 hour, 3 hours, and 6 hours), so as to better understand the changing trend of air quality in the future.

[0064] If the SVR model predicts that air quality may deteriorate in the future (such as an increase in pollutant concentration), the system will increase the operating intensity of the air purification module in advance, adjust the airflow direction to control the spread of pollution, and appropriately adjust the temperature and humidity to reduce the impact of pollution. If the forecast shows that air quality will remain good in the future (such as a decrease in pollutant concentration), the system will reduce the power of the air purification module and reduce the frequency of airflow and temperature and humidity control to save energy.

[0065] In this embodiment, the air quality data (such as pollutant concentration, temperature and humidity, air flow velocity, etc.) of each experimental area are first collected from the sensor module, the missing values ​​and outliers in the data are processed, and the feature input (such as the mean and maximum values ​​of the pollutant concentration in the past 24 hours) is generated. Then the feature data is standardized and divided into a training set and a test set in chronological order. The radial basis kernel function (RBF) of the support vector regression (SVR) model is selected to process nonlinear data, and the model parameters are optimized through grid search and cross-validation to maintain the best prediction effect under different settings. After the model is trained using the training set, the prediction error is evaluated on the test set to ensure the accuracy of the model. In actual operation, the system inputs the latest air quality data to predict the trend of air quality changes in the future. If the forecast shows that the air quality will deteriorate, the system will increase the purification intensity, adjust the airflow direction and temperature and humidity in advance; if the forecast remains good, the purification and adjustment frequency will be reduced to optimize energy consumption.

[0066] 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.

[0067] 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.

[0068] 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.

[0069] 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. A laboratory air conditioning system, characterized in that: It includes a central control module, a regional sensor module, a partition control module, an air purification module and an airflow control module: the central control module is used to control the entire system and communicate with each partition control module to achieve linkage control of multi-region air quality parameters; the regional sensor module includes a temperature and humidity sensor, an air quality sensor and an airflow velocity sensor, which are distributed in various experimental areas of the laboratory and are used to detect and feedback the air quality, temperature and humidity and airflow data of each area in real time; the partition control module is located in each experimental area of ​​the laboratory and is used to judge the severity of air pollution in each experimental area according to the data fed back by the regional sensor module, and dynamically adjust the air purification module, temperature and humidity adjustment module and airflow control module of the experimental area according to the severity of air pollution to achieve independent air quality control; The air purification module, including a high-efficiency air filter, an activated carbon filter and an ultraviolet disinfection device, is used to remove volatile organic compounds, bioaerosols and particulate matter in the air; the airflow control module is used to adjust the airflow speed and direction of each area according to the changes in the severity of air pollution in each experimental area within a fixed time period, to ensure the orderly flow of air in each area and avoid the spread of pollutants.

2. A laboratory air conditioning system according to claim 1, characterized in that: In the partition control module, the pollutant concentration fluctuation index is generated after analyzing the abnormal fluctuation of the time change rate of the pollutant concentration in each experimental area. The method for obtaining the pollutant concentration fluctuation index is: Set a time window x, that is, the number of data points included in the calculation of the volatility index, assign a weight wi to each data point in the given time window, and mark the pollutant concentration at x time points as ;in, is the earliest data point, For the latest data point, calculate the weighted moving average EF of the pollutant concentration, the expression is: ; In the formula, is the pollutant concentration at the i-th time point, and the pollutant concentration fluctuation index is defined as the latest actual concentration value The difference with EF is expressed as: ; In the formula, is the pollutant concentration fluctuation index.

3. A laboratory air conditioning system according to claim 2, characterized in that: In the partition control module, the air flow velocity sensor data is used to monitor the air flow pattern in the area, and the air flow diffusion interference index is generated according to the interference degree of the air flow velocity and direction changes. The method for obtaining the air flow diffusion interference index is: Obtain the time series data of airflow velocity and direction from the airflow velocity sensor, set the sampling frequency f, collect the data for the past T seconds, and select a time series representing the steady state of airflow from the historical data as the baseline airflow pattern, recorded as ;in Represents the speed and direction data at the nth moment in the reference airflow pattern, and records the sequence data of the airflow speed and direction in the current time period, recorded as ,in represents the speed and direction data at the mth moment in the current airflow pattern. Assuming that each data point contains the airflow speed v and the airflow direction θ, then the i-th moment reference point With the current airflow point The Euclidean distance between for: ; Construct an n×m DTW matrix D, where D(i,j) represents the reference airflow pattern With the current airflow model The cumulative minimum matching cost is initialized to D(0,0)=0. For each i∈[1,n] and j∈[1,m], the recursive formula is used to calculate each element in the DTW matrix: ; The airflow diffusion interference index is defined as the final matching cost between the reference airflow pattern and the current airflow pattern, and the expression is: ; Where SD is the airflow diffusion interference index, D(n,m) is the final element of the matrix D, and D(n,m) represents the minimum matching cost of the two airflow patterns.

4. A laboratory air conditioning system according to claim 3, characterized in that: The pollutant concentration fluctuation index and the airflow diffusion interference 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 predicts the air pollution severity value label in each experimental area with each group of first eigenvectors as the prediction target, and minimizes the sum of prediction errors of the air pollution severity value labels in all experimental areas as the training target. The machine learning model is trained until the sum of prediction errors converges and the model training is stopped. The air pollution severity value in each experimental area is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

5. A laboratory air conditioning system according to claim 4, characterized in that: The acquired air pollution severity value in each experimental 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 air pollution severity value in each experimental area is compared with the first standard threshold and the second standard threshold respectively; If the air pollution severity value in the experimental area is greater than the second standard threshold, it means that the air pollution severity in the experimental area is high. At this time, a high pollution degree signal is generated, and the experimental area is divided into a high pollution area; If the air pollution severity value in the experimental 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 air pollution severity in the experimental area is average. At this time, a medium pollution degree signal is generated and the experimental area is divided into a medium pollution area; If the air pollution severity value in the experimental area is less than the first standard threshold, it means that the air pollution severity in the experimental area is low. At this time, a low pollution degree signal is generated and the experimental area is divided into a low pollution area.

6. A laboratory air conditioning system according to claim 5, characterized in that: In the airflow control module, the airflow velocity v is set to control the diffusion trend of pollutants. The airflow velocity should be proportional to the rate of change of the severity of air pollution S. When the severity of pollution increases, the airflow velocity is increased to quickly remove pollutants; when the severity of pollution is low or stable, the airflow velocity is maintained at a low level to save energy. The specific formula is: ; is the airflow velocity at the current time t, is the basic airflow velocity used to maintain air flow and prevent pollutant accumulation. is the airflow velocity adjustment factor, which is used to control the sensitivity of the impact of changes in pollution severity on airflow velocity. is the rate of change of air pollution severity S; The airflow angle θ is set to deviate from the pollution source area. The airflow direction adjustment formula is: ; The airflow direction angle at the current time t, is the angle of the target exhaust direction, is the airflow direction adjustment coefficient, which is used to control the adjustment range of airflow direction due to pollution severity. is the change in the severity of air pollution.

7. A laboratory air conditioning system according to claim 6, characterized in that: In the air flow control module, when >0 and ΔS>0, it means that the pollution is aggravated and the air flow speed should be increased rapidly And adjust the airflow direction Close to the exhaust system to avoid the spread of pollutants; <0 and ΔS<0, indicating that pollution is alleviated, and the air flow speed is gradually reduced. To base speed And restore the original direction to maintain the stability of regional air flow.

8. A laboratory air conditioning control method, used to implement a laboratory air conditioning system according to any one of claims 1 to 7, characterized in that: S1: Monitor the air quality, temperature, humidity and airflow data of the experimental area through the regional sensor module, and feed the data back to the central control module; S2: Analyze the severity of air pollution in each experimental area based on the detection data, and dynamically adjust the air purification module, temperature and humidity adjustment module and airflow control module in each area; S3: When it is detected that the severity of air pollution in the experimental area changes abnormally, the airflow speed and direction of each experimental area are adjusted in real time; S4: Predict the air quality change trend in each experimental area, and adjust the air quality parameters in advance according to the prediction results to optimize energy consumption.

9. A laboratory air conditioning control method according to claim 8, characterized in that: In S4, air quality data of each experimental area is collected, missing values ​​and outliers in the data are processed, and feature inputs are generated. The feature data are then standardized and divided into training and test sets in chronological order. The radial basis kernel function of the support vector regression model is used to process nonlinear data. The model parameters are optimized through grid search and cross-validation to maintain the best prediction effect under different settings. After the model is trained using the training set, the latest air quality data is input to predict the trend of air quality changes in the future. If the prediction shows that the air quality will deteriorate, the purification intensity will be increased in advance, and the airflow direction and temperature and humidity will be adjusted. If the prediction remains good, the purification and adjustment frequency will be reduced to optimize energy consumption.

Images