Gas environment regulation and control system and method for clean laboratory
By designing a clean laboratory gas environment regulation system including a three-stage filtration module, an activated carbon adsorption module and a multi-dimensional sensor network, the problems of insufficient purification capacity, lag in control accuracy and high energy consumption in traditional systems are solved, and efficient, accurate and energy-saving gas environment control is achieved.
Patent Information
- Application Number
- CN202510562833.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-30
- Publication Date
- 2025-06-17
AI Technical Summary
The gas environment regulation system of traditional clean laboratories has problems such as insufficient purification capacity, lagging control accuracy, high energy consumption and isolation of data, and cannot effectively deal with sudden pollution leakage or load changes during the experiment.
A gas environment regulation system including an air treatment unit, a gas monitoring unit and an intelligent control unit is designed. The air treatment unit adopts a three-stage filtration module and an activated carbon adsorption module, combining a variable refrigerant flow air conditioning system and a heat recovery device. The gas monitoring unit uses a multi-dimensional sensor network and a movable sampling device, and the intelligent control unit performs real-time monitoring and optimization through edge computing and cloud management platforms.
It realizes efficient purification of particulate matter, microorganism and chemical pollutants, improves the accuracy and response speed of gas environment control, reduces energy consumption, and improves the flexibility and management efficiency of the system through intelligent and modular design.
Smart Images

Figure CN120160272A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of clean room environment control, and particularly relates to a gas environment regulation system and method for a clean laboratory. Background Art
[0002] Clean laboratories have important applications in fields such as biomedicine and semiconductor manufacturing. The stability of their gas environment directly affects the accuracy of experimental results and personnel safety. In modern scientific research and industrial production, the quality of the gas environment in clean laboratories directly affects the accuracy of experimental results, the yield of products, and personnel safety. The traditional regulation system has the following technical bottlenecks:
[0003] Insufficient purification ability: A single filtration level is difficult to simultaneously remove particulate matter (such as 0.1μm ultrafine particles), microorganisms (such as bacterial spores), and chemical pollutants (such as VOCs). Moreover, after the activated carbon is saturated with adsorption, it needs to be replaced manually, resulting in high maintenance costs;
[0004] Lag in control accuracy: The temperature, humidity, and air flow distribution rely on manual experience for adjustment, with a long response time (typically >5 minutes), and cannot cope with sudden pollution leaks or load changes during the experiment;
[0005] High energy consumption: The traditional constant air volume air conditioning system and fixed filtration mode lead to energy waste. The annual energy consumption per unit area of some laboratories exceeds 150 kWh / m 2 ;
[0006] Data isolation problem: The sensor data does not form a closed-loop control. In case of abnormal environmental parameters, there is a lack of automatic response, and it cannot be linked with the laboratory information management system (LIMS).
[0007] In the prior art, Chinese Patent CN209801234U discloses a laboratory ventilation system, but it only realizes simple regulation through single filtration and a variable frequency fan, and does not solve the problems of multi-pollutant collaborative purification and intelligent decision-making; US Patent US10234567B2 proposes laboratory environment control based on a sensor network, but lacks dynamic simulation of the air flow field and energy consumption optimization strategies. Therefore, there is an urgent need for a gas environment regulation system with high precision, low energy consumption, and intelligence. Summary of the Invention
[0008] The purpose of the present invention is to provide a gas environment regulation system and method for a clean laboratory to solve the problems raised in the above background art.
[0009] To achieve the above object, the present invention provides the following technical solution: A gas environment control system for a clean laboratory, comprising an air treatment unit and a gas monitoring unit, wherein the air treatment unit and the gas monitoring unit are respectively connected to an intelligent control unit; the control unit is connected to an actuator; the air treatment unit includes a three-stage filtration module, an activated carbon adsorption module, a variable refrigerant flow air conditioning system and a heat recovery device; the gas monitoring unit includes a multi-dimensional sensor network, a movable sampling device and an orbital suction hood; the intelligent control unit includes an edge computing node and a cloud management platform; the actuator includes an air flow control component and a temperature and humidity regulation component.
[0010] Preferably, the three-stage filtration module is sequentially connected in series along the air flow direction. The three-stage filtration module includes a primary filter, a medium filter and a high-efficiency filter; the primary filter is a non-woven fabric filter screen, the medium filter is a glass fiber filter screen, and the high-efficiency filter is a HEPA filter screen; the primary filter, the medium filter and the high-efficiency filter are connected by a quick-release flange interface; the activated carbon adsorption module is arranged downstream of the high-efficiency filter, and the activated carbon adsorption module includes a honeycomb activated carbon unit and a thermal regeneration device; the variable refrigerant flow air conditioning system includes a DC variable frequency compressor, a plate heat exchanger, an electronic expansion valve and an indoor unit, and the indoor unit is connected to the outdoor unit through an R410A refrigerant pipeline; the heat recovery device adopts a plate sensible heat exchanger, and the plate sensible heat exchanger is arranged at the intersection of the fresh air and exhaust air pipelines and the fresh air and exhaust air are arranged in a cross-flow manner.
[0011] Preferably, the multi-dimensional sensor network includes a PID sensor, a PM2.5 sensor, a bioaerosol sensor and a dynamic pressure sensor. The PID sensor, the PM2.5 sensor, the bioaerosol sensor and the dynamic pressure sensor are respectively connected to a data acquisition module through an RS-485 bus; the movable sampling device is an orbital suction hood, and the orbital suction hood is driven by an X-Y axis linear guide rail and is equipped with a three-stage telescopic adjustment arm; the sampling path planning module generates an optimal sampling path based on the Dijkstra algorithm.
[0012] Preferably, the edge computing node adopts an ARM Cortex-A9 processor, integrates Modbus / TCP and MQTT protocol stacks, and processes sensor data in real time; the cloud management platform includes: a time series database, a remote monitoring interface and a LIMS system interface; the LIMS system interface realizes the automatic association of environmental data and experimental records through a RESTful API; the air flow control component includes a variable air volume Venturi valve, a constant air volume valve and an intelligent damper electric actuator; the temperature and humidity regulation component includes an electrode type humidifier and a rotary dehumidifier, and the electrode type humidifier and the rotary dehumidifier are connected in parallel to the air conditioning supply air pipeline and the output power is adjusted by a proportional integral derivative controller.
[0013] Preferably, the three - stage filtration module and the activated carbon adsorption module are integrated through an aluminum alloy frame, and the joints of the aluminum alloy frame are sealed with silicone rubber; a differential pressure sensor is provided between the primary filter and the intermediate filter, and a replacement alarm is triggered when the differential pressure exceeds a preset value.
[0014] Preferably, the track - type suction hood of the gas monitoring unit is powered by a sliding contact line, and a six - axis force sensor is provided at the connection between the track - type suction hood and the adjusting arm; the edge computing node of the intelligent control unit is connected to the cloud platform through a 5G communication module, and the data encryption in the 5G communication module adopts the AES - 256 protocol; the variable air volume venturi valve of the actuator is connected to the intelligent damper electric actuator through a Profibus - DP bus.
[0015] The present invention also discloses a method for regulating the gas environment in a clean laboratory, including the following steps:
[0016] Step 1: Multi - dimensional data collection and pre - processing:
[0017] Spatial grid sampling: The track - type suction hood moves along a preset path, stops for 5 seconds every 1 m to collect gas samples, and generates a three - dimensional concentration distribution matrix of the laboratory.
[0018] Data noise reduction processing: The Kalman filter algorithm is used to reduce the noise of the sensor data, remove high - frequency noise, and perform threshold verification.
[0019] Step 2: Intelligent analysis and strategy generation:
[0020] Three - dimensional air flow simulation: Input the geometric model of the laboratory, which includes obstacles such as fume hoods and experimental tables, the positions of supply and exhaust air outlets, and the current temperature and humidity parameters, calculate the air flow field, and predict the time threshold for pollutants to spread to the buffer zone.
[0021] Reinforcement learning optimization: Construct a state space S, an action space A, and a reward function R, and generate an optimal regulation strategy through the Q - learning algorithm; the state space S includes temperature and humidity, pressure, and pollutant concentration; the action space A includes supply and exhaust air volumes, filtration levels, and temperature and humidity set values; the reward function R = 0.1×energy consumption−0.9×∑|parameter deviation|.
[0022] Step 3: Dynamic regulation and execution:
[0023] Filtration system control: When the PM2.5 concentration > 10 μg / m 3 ³, the primary, intermediate, and high - efficiency filtrations are started in sequence; when the TVOC concentration > 10 ppm and the operation time of the activated carbon adsorption module > 4 hours, a thermal regeneration program is triggered.
[0024] Airflow and Pressure Control: Based on the CFD simulation results, the flow rates of each supply and exhaust air outlet are adjusted through variable air volume Venturi valves to maintain the average cross-sectional wind speed in the clean area at 0.3 - 0.5 m / s; when the dynamic pressure sensor detects that the pressure difference between the clean area and the buffer area deviates from -35 Pa ± 2 Pa, the pressure compensation algorithm is activated;
[0025] Temperature and Humidity Regulation: The fuzzy PID control algorithm is adopted, with the temperature set value at 20 - 25 °C and the humidity set value at 40% - 60% RH. The supply air temperature is dynamically adjusted according to the density of laboratory personnel;
[0026] Step Four: Remote Monitoring and System Maintenance:
[0027] Real-time Warning Mechanism: When the concentration of bioaerosol > 100 CFU / m 3 or the concentration of CO2 > 1500 ppm, the cloud platform will send an audible and visual alarm and a text message to notify the responsible person, and automatically generate a maintenance work order;
[0028] Energy Consumption Management Module: Based on the ARIMA model, the energy consumption of the next day is predicted, and the compressor frequency and fresh air ratio of the VRF air conditioning system are optimized.
[0029] Preferably, the time series database updates the flow field data every 30 minutes. When a vortex area is detected, the angle of the air supply outlet is automatically adjusted.
[0030] Compared with the prior art, the beneficial effects of the present invention are:
[0031] The gas environment control system of this clean laboratory adopts a composite purification technology, with the function of real-time monitoring and purification; this system adopts a three-stage filtration module, and the filtration efficiency for particulate matter covers the particle size range from 5 μm to 0.1 μm. Combined with an activated carbon thermal regeneration device, it realizes the cyclic treatment of chemical pollutants and reduces the frequency of consumable replacement by more than 50%; the heat recovery device recovers the energy of the exhaust air, reduces the energy consumption of fresh air heating / cooling, and can reduce the air conditioning system load by 25% through calculation. This system adopts dynamic perception and intelligent decision-making: the orbital suction hood realizes grid sampling of the laboratory space. Cooperating with the bioaerosol sensor, it can detect microbial particles above 0.3 μm, meeting the requirements of a biosafety laboratory (BSL-2); the reinforcement learning algorithm optimizes the control strategy in real time. Compared with the traditional PID control, the response speed is increased by 30%, and the parameter fluctuation range is reduced by 40%. This system also adopts modular and intelligent integration: each module is connected through standardized interfaces (flanges, bus protocols), supporting quick disassembly and expansion; the cloud platform realizes remote monitoring of equipment, data traceability, and integration with the LIMS system, meeting the ISO17025 laboratory management standard. Description of the Drawings
[0032] Figure 1One of the system block diagrams of the present invention;
[0033] Figure 2 Another system block diagram of the present invention. Detailed implementation manners
[0034] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.
[0035] Embodiment 1
[0036] Refer to Figure 1 and Figure 2 , a gas environment control system for a clean laboratory, including an air treatment unit and a gas monitoring unit, wherein the air treatment unit and the gas monitoring unit are respectively connected to an intelligent control unit; the control unit is connected to an actuator; the air treatment unit includes a three-stage filtration module, an activated carbon adsorption module, a variable refrigerant flow air conditioning system and a heat recovery device; the gas monitoring unit includes a multi-dimensional sensor network, a movable sampling device and an orbital suction hood; the intelligent control unit includes an edge computing node and a cloud management platform; the actuator includes an air flow control component and a temperature and humidity control component.
[0037] Among them, the three-stage filtration module is sequentially connected in series along the air flow direction. The three-stage filtration module includes a primary filter, a medium filter and a high-efficiency filter; the primary filter is a non-woven fabric filter screen, the medium filter is a glass fiber filter screen, and the high-efficiency filter is a HEPA filter screen; the primary filter, the medium filter and the high-efficiency filter are connected by a quick-release flange interface; the activated carbon adsorption module is arranged downstream of the high-efficiency filter, and the activated carbon adsorption module includes a honeycomb activated carbon unit and a heat regeneration device; the variable refrigerant flow air conditioning system includes a DC variable frequency compressor, a plate heat exchanger, an electronic expansion valve and an indoor unit, and the indoor unit is connected to the outdoor unit through an R410A refrigerant pipeline; the heat recovery device adopts a plate sensible heat exchanger, and the plate sensible heat exchanger is arranged at the intersection of the fresh air and exhaust air pipelines and the fresh air and exhaust air are arranged in a cross-flow manner.
[0038] Among them, the multi-dimensional sensor network includes a PID sensor, a PM2.5 sensor, a bio-aerosol sensor, and a dynamic pressure sensor. The PID sensor, PM2.5 sensor, bio-aerosol sensor, and dynamic pressure sensor are respectively connected to the data acquisition module through the RS-485 bus; the movable sampling device is an orbital suction hood, and the orbital suction hood is driven by an X-Y axis linear guide and is equipped with a three-stage telescopic adjustment arm; the sampling path planning module generates an optimal sampling path based on the Dijkstra algorithm,
[0039] Among them, the edge computing node uses an ARM Cortex-A9 processor, integrates Modbus / TCP and MQTT protocol stacks, and processes sensor data in real time; the cloud management platform includes: a time series database, a remote monitoring interface, and a LIMS system interface; the LIMS system interface realizes the automatic association of environmental data and experimental records through the RESTful API; the air flow control component includes a variable air volume Venturi valve, a constant air volume valve, and an intelligent damper electric actuator; the temperature and humidity control component includes an electrode humidifier and a rotary dehumidifier. The electrode humidifier and the rotary dehumidifier are connected in parallel to the air conditioning supply air pipeline, and the output power is adjusted by a proportional integral derivative controller.
[0040] Among them, the three-stage filtration module and the activated carbon adsorption module are integrated through an aluminum alloy frame, and the joints of the aluminum alloy frame are sealed with silicone rubber; the primary filter and the medium filter are provided with differential pressure sensors, and when the differential pressure exceeds the preset value, a replacement alarm is triggered.
[0041] Among them, the orbital suction hood of the gas monitoring unit is powered by a sliding contact wire, and a six-axis force sensor is arranged at the connection between the orbital suction hood and the adjustment arm; the edge computing node of the intelligent control unit is connected to the cloud platform through a 5G communication module, and the data encryption in the 5G communication module uses the AES-256 protocol; the variable air volume Venturi valve of the actuator is connected to the intelligent damper electric actuator through a Profibus-DP bus.
[0042] This embodiment also discloses a gas environment regulation method for a clean laboratory, including the following steps:
[0043] Step 1. Multi-dimensional data collection and preprocessing:
[0044] Spatial grid sampling: The orbital suction hood moves according to a preset path, stops for 5 seconds every 1 m to collect gas samples, and generates a three-dimensional concentration distribution matrix of the laboratory;
[0045] Data noise reduction processing: The Kalman filter algorithm is used to reduce the noise of the sensor data, remove high-frequency noise, and perform threshold verification;
[0046] Step 2: Intelligent Analysis and Strategy Generation:
[0047] Three-dimensional airflow simulation: Input the laboratory geometric model, which includes obstacles such as fume hoods and experimental benches, the positions of supply and exhaust air vents, and current temperature and humidity parameters, calculate the airflow field, and predict the time threshold for pollutants to spread to the buffer zone;
[0048] Reinforcement learning optimization: Construct a state space S, an action space A, and a reward function R, and generate an optimal control strategy through the Q-learning algorithm; the state space S includes temperature and humidity, pressure, and pollutant concentration; the action space A includes supply and exhaust air volumes, filtration levels, and temperature and humidity set values; the reward function R = 0.1×energy consumption - 0.9×∑|parameter deviation|;
[0049] Step 3: Dynamic Regulation and Execution:
[0050] Filtration system control: When the PM2.5 concentration > 10 μg / m 3 , start primary, intermediate, and high-efficiency filtration in sequence; when the TVOC concentration > 10 ppm and the operation time of the activated carbon adsorption module > 4 hours, trigger the thermal regeneration program;
[0051] Airflow and pressure control: Based on the CFD simulation results, adjust the flow rates of each supply and exhaust air vent through variable air volume Venturi valves to maintain the average cross-sectional wind speed in the clean area at 0.3 - 0.5 m / s; when the dynamic pressure sensor detects that the pressure difference between the clean area and the buffer zone deviates from -35 Pa ± 2 Pa, start the pressure compensation algorithm;
[0052] Temperature and humidity regulation: Adopt the fuzzy PID control algorithm, with the temperature set value of 20 - 25 °C and the humidity set value of 40% - 60% RH, and dynamically adjust the supply air temperature according to the personnel density in the laboratory;
[0053] Step 4: Remote Monitoring and System Maintenance:
[0054] Real-time warning mechanism: When the bioaerosol concentration > 100 CFU / m 3 or the CO2 concentration > 1500 ppm, the cloud platform will notify the responsible person through audible and visual alarms and text messages, and automatically generate a maintenance work order;
[0055] Energy consumption management module: Predict the next day's energy consumption based on the ARIMA model, and optimize the compressor frequency and fresh air ratio of the VRF air conditioning system.
[0056] Among them, the time series database updates the flow field data every 30 minutes, and when a vortex area is detected, the air supply outlet angle is automatically adjusted.
[0057] Among them, the gas environment control system of this clean laboratory adopts a composite purification technology and has the function of real-time monitoring and purification; this system adopts a three-stage filtration module, and the filtration efficiency of particulate matter covers the particle size range from 5μm to 0.1μm. Combined with an activated carbon thermal regeneration device, it realizes the cyclic treatment of chemical pollutants and reduces the consumable replacement frequency by more than 50%; the heat recovery device recovers the exhaust air energy, reduces the fresh air heating / cooling energy consumption, and can reduce the air-conditioning system load by 25% as measured. This system adopts dynamic perception and intelligent decision-making: the orbital suction hood realizes grid sampling of the laboratory space. Cooperating with a bio-aerosol sensor, it can detect microbial particles larger than 0.3μm and meets the requirements of a biosafety laboratory (BSL-2); the reinforcement learning algorithm optimizes the control strategy in real time. Compared with the traditional PID control, the response speed is increased by 30%, and the parameter fluctuation range is reduced by 40%. This system also adopts modular and intelligent integration: each module is connected through standardized interfaces (flanges, bus protocols), supporting rapid disassembly and expansion; the cloud platform realizes remote monitoring of equipment, data traceability, and LIMS system integration, meeting the ISO17025 laboratory management standard.
[0058] Example 2
[0059] Please refer to Figure 1 and Figure 2 , this example provides a gas environment control system for a clean laboratory, which mainly includes the following parts:
[0060] I. Air handling unit, including the following parts:
[0061] 1. Three-stage filtration module
[0062] Primary filter: It uses a G4-grade non-woven fabric filter screen, the frame is made of aluminum alloy, the rated air volume is 2000m 3 / h, the pressure loss ≤ 150Pa, the filtration efficiency of particles ≥ 5μm is ≥ 80%, and a differential pressure sensor (model: E+E EE870) is set to monitor the clogging state of the filter screen in real time;
[0063] Intermediate filter: F8-grade fiberglass filter screen, the folded structure increases the filtration area (15m 2 ), the rated air volume is 1500m 3 / h, the pressure loss ≤ 250Pa, the filtration efficiency of particles ≥ 1μm is ≥ 95%;
[0064] High-efficiency filter: H13-grade HEPA filter screen, using ultra-fine fiberglass filter paper, the rated air volume is 1000m 3 / h, the pressure loss ≤ 400Pa, the filtration efficiency of 0.1μm particles is ≥ 99.97% (tested by the DOP method). 2. Activated carbon adsorption and thermal regeneration
[0065] Adsorption module: honeycomb activated carbon unit (pore diameter 1 mm, bulk density 450 kg / m 3 ), total adsorption capacity ≥ 20 kg, the main path and the regeneration branch are switched through a pneumatic butterfly valve (model: ASCO 8262G);
[0066] Thermal regeneration device: an electric heating element (nickel-chromium alloy wire, power 10 kW) heats the air to 850 °C, equipped with a steam generator (flow rate 0.5 m 3 / h), regeneration time 45 minutes, and the adsorption efficiency of the activated carbon returns to over 90% after regeneration.
[0067] 3. VRF air conditioning system
[0068] DC variable frequency compressor (Mitsubishi Electric JT160), frequency conversion range 15 - 120 Hz, energy efficiency ratio (EER) ≥ 3.5;
[0069] The electronic expansion valve (Danfoss EVR32) adjusts the opening in real time according to the superheat degree, and cooperates with the plate heat exchanger (Alfa Laval M10 - B) to achieve precise refrigeration / heating, with a temperature control accuracy of ±0.3 °C.
[0070] 4. Heat recovery device
[0071] Plate sensible heat exchanger (aluminum core, heat transfer coefficient 50 W / (m 2 ·K)), the fresh air and the exhaust air flow in opposite directions, recovering 70% of the sensible heat energy of the exhaust air, and reducing the fresh air heating energy consumption by 40% in winter.
[0072] II. Gas monitoring unit, including the following parts:
[0073] 1. Sensor configuration
[0074] PID sensor: Honeywell Solstice AQ7, detection principle is photoionization, response time < 15 seconds, linearity ±2%;
[0075] Bioaerosol sensor: German Grimm 1.109, using laser-induced fluorescence technology, can distinguish biological and non-biological particles, detection limit 5 CFU / m 3 ;
[0076] Dynamic pressure sensor: American Validity VTI - 820, range ±100 Pa, accuracy ±0.25 Pa, with 1 measuring point arranged at each of the four corners and the center position in the clean area.
[0077] 2. Movable sampling device
[0078] Track system: The X-axis adopts a double-rail design (load-bearing 20 kg), the Y-axis is a single rail, the motor is a stepper motor (step angle 1.8°), and the positioning accuracy is ±1 mm;
[0079] Suction hood: Made of polytetrafluoroethylene, with an inlet air velocity of 1.5 m / s and an in-built flow sensor (Omega FMA-1800) to monitor the sampling flow rate in real time.
[0080] III. Intelligent control unit, including the following parts:
[0081] 1. Edge computing node:
[0082] Hardware platform: Advantech UNO-2184G, equipped with an ARM Cortex-A9 quad-core processor, integrated with 1GB RAM and 8GB Flash;
[0083] Software architecture: Linux operating system, running OpenFOAM 8.0 for CFD simulation, with each simulation taking less than 2 minutes (mesh number: 500,000), and the results are used to guide the adjustment of the air supply outlet angle.
[0084] 2. Cloud platform:
[0085] Database: InfluxDB cluster deployment, supporting 1000 data writes per second, and historical data can be visually analyzed through Grafana;
[0086] Alarm mechanism: Set three levels of alarms (early warning / alarm / emergency), corresponding to different notification methods (APP push / sms / acoustic and optical alarm). For example, when the concentration of bioaerosol > 200 CFU / m 3 an emergency shutdown is triggered.
[0087] IV. Actuator, including the following parts:
[0088] 1. Airflow control components:
[0089] Variable air volume Venturi valve: Pressure-independent design, flow control accuracy of ±5%, response time < 2 seconds;
[0090] Intelligent damper electric actuator: Siemens SQX62, with a position feedback potentiometer, supporting 0-10V signal input, and the opening control accuracy is ±1%.
[0091] 2. Temperature and humidity regulation:
[0092] Electrode humidifier: Power 3kW, humidification capacity 5kg / h, response time < 30 seconds;
[0093] Rotary dehumidifier: Munters MD-30, dehumidification capacity 30kg / h (inlet humidity 80% RH), regeneration temperature 120°C, energy consumption 1.5kW.
[0094] V. Detailed working process is as follows:
[0095] Self-cleaning startup (initialization phase):
[0096] After the system is powered on, the air treatment unit runs at the maximum air volume (2000 m 3 / h) for 30 minutes to complete the self-cleaning of the laboratory;
[0097] The orbital suction hood automatically returns to the origin, and the sensor performs zero calibration (calibration period: 7 days).
[0098] Normalized monitoring (operation phase):
[0099] The sensor collects data every 10 seconds and inputs it into the edge computing node after Kalman filtering;
[0100] The CFD module updates the airflow simulation results every hour, identifies potential vortex regions (such as behind the experimental bench), and adjusts the corresponding air supply outlet angle (5° each time).
[0101] Pollution response (abnormality handling):
[0102] When the TVOC concentration > 50 ppm (such as reagent leakage), the system automatically switches to the "emergency mode": closes the activated carbon adsorption bypass valve, fully opens the fresh air valve (fresh air ratio 100%), and simultaneously starts the plasma purifier (treatment air volume 2000 m 3 / h);
[0103] If the bioaerosol concentration > 100 CFU / m 3 , trigger the ultraviolet germicidal lamp (wavelength 254 nm, irradiation intensity 150 μW / cm 2 ), irradiate for 30 minutes, and simultaneously send a pollution event record to the LIMS system.
[0104] Energy consumption optimization (non-working period):
[0105] When the personnel positioning system (RFID tag) detects that there is no one in the laboratory, start the "energy-saving mode": reduce the air supply volume to 50%, adjust the air conditioner temperature set value to 28°C (summer) / 18°C (winter), and extend the activated carbon regeneration period to 8 hours.
[0106] VI. Experimental verification. The experimental verification results of this system are as follows:
[0107] Test was carried out in a semiconductor wafer manufacturing laboratory (area 500 m 2 , cleanliness class ISO 6) to compare the performance of the traditional system and the system of the present invention:
[0108] Test Items Traditional System System of the Present Invention Improvement Range 0.1μm Particle Filtration Efficiency 99.5% 99.98% +0.4% TVOC Removal Rate 80% 95% +15% Temperature Control Precision ±1.0℃ ±0.3℃ +70% Pressure Gradient Stability ±5Pa ±1Pa +80% Energy Consumption (kWh / day) 288 192 -33.3% Pollution Response Time 300 seconds 60 seconds -80%
[0109] It can be found from the experimental results that the beneficial effects of the present invention include:
[0110] Multi-pollutant collaborative purification: The three-stage filtration and activated carbon thermal regeneration technology can efficiently treat particulate matter, microorganisms, and chemical pollutants, meeting the ISO 14644-1 and GB 50346 biosafety laboratory standards;
[0111] Precise dynamic control: Based on CFD simulation and reinforcement learning algorithm, it can achieve high-precision control of temperature and humidity within ±0.3°C / ±2%RH and pressure gradient within ±2Pa, which is 30% better than the industry standard;
[0112] Intelligent energy-saving and consumption reduction: The combination of heat recovery device and VRF air-conditioning system can reduce energy consumption by 33.3%, saving about 150,000 yuan in electricity bills annually (calculated at 0.8 yuan / kWh);
[0113] Full-process intelligentization: The full-closed-loop automation from data collection, strategy generation to equipment control can reduce manual intervention by 90% and improve the laboratory management efficiency.
[0114] The present invention also has industrial applicability. The clean laboratory gas environment regulation system and method provided by the present invention can be widely applied to high-end clean laboratories in fields such as biomedicine, semiconductors, and new materials. It has mature hardware interfaces and software protocols, supports customized development and rapid deployment. The system adopts a modular design, and each component can be independently replaced and maintained to adapt to the environmental control requirements of laboratories of different scales, with significant economic benefits and technological advancement.
[0115] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A gas environment control system for a clean laboratory, characterized in that: It includes an air handling unit and a gas monitoring unit, which are respectively connected to an intelligent control unit; the control unit is connected to an actuator; the air handling unit includes a three-stage filtration module, an activated carbon adsorption module, a variable refrigerant flow air-conditioning system and a heat recovery device; the gas monitoring unit includes a multi-dimensional sensor network, a movable sampling device and a track-type suction hood; the intelligent control unit includes an edge computing node and a cloud management platform; the actuator includes an airflow control component and a temperature and humidity control component.
2. The gas environment control system for a clean laboratory according to claim 1 is characterized in that: The three-stage filtration modules are arranged in series in sequence along the air flow direction, and the three-stage filtration modules include a primary filter, a medium filter, and a high-efficiency filter; the primary filter is a non-woven filter, the medium filter is a glass fiber filter, and the high-efficiency filter is a HEPA filter; the primary filter, the medium filter, and the high-efficiency filter are connected by a quick-release flange interface; the activated carbon adsorption module is arranged downstream of the high-efficiency filter, and the activated carbon adsorption module includes a honeycomb activated carbon unit and a heat regeneration device; the variable refrigerant flow air-conditioning system includes a DC variable frequency compressor, a plate heat exchanger, an electronic expansion valve and an indoor unit, and the indoor unit is connected to the outdoor unit through an R410A refrigerant pipeline; the heat recovery device adopts a plate sensible heat exchanger, and the plate sensible heat exchanger is arranged at the intersection of the fresh air and exhaust air pipelines, and the fresh air and exhaust air are arranged in a cross flow.
3. The gas environment control system for a clean laboratory according to claim 1, characterized in that: The multidimensional sensor network includes a PID sensor, a PM2.5 sensor, a bioaerosol sensor, and a dynamic pressure sensor, and the PID sensor, PM2.5 sensor, bioaerosol sensor, and dynamic pressure sensor are respectively connected to a data acquisition module via an RS-485 bus; the movable sampling device is a track-type suction hood, which is driven by an XY-axis linear guide and equipped with a three-stage telescopic adjustment arm; the sampling path planning module generates an optimal sampling path based on a Dijkstra algorithm.
4. The gas environment control system for a clean laboratory according to claim 1, characterized in that: The edge computing node adopts an ARM Cortex-A9 processor, integrates Modbus / TCP and MQTT protocol stacks, and processes sensor data in real time; the cloud management platform includes: a time series database, a remote monitoring interface and a LIMS system interface; the LIMS system interface realizes automatic association of environmental data and experimental records through a RESTful API; the airflow control component includes a variable air volume Venturi valve, a constant air volume valve and an intelligent damper electric actuator; the temperature and humidity control component includes an electrode humidifier and a rotary dehumidifier, the electrode humidifier and the rotary dehumidifier are connected in parallel to the air supply pipeline of the air conditioner, and the output power is adjusted by a proportional integral differential controller.
5. The gas environment control system for a clean laboratory according to claim 1, characterized in that: The three-stage filtration module and the activated carbon adsorption module are integrated through an aluminum alloy frame, and the joints of the aluminum alloy frame are sealed with silicone rubber; the primary filter and the medium filter are provided with a pressure difference sensor, and a replacement alarm is triggered when the pressure difference exceeds a preset value.
6. The gas environment control system for a clean laboratory according to claim 1, characterized in that: The track-type air suction hood of the gas monitoring unit is powered by a busbar, and a six-axis force sensor is arranged at the connection between the track-type air suction hood and the adjusting arm; the edge computing node of the intelligent control unit is connected to the cloud platform through a 5G communication module, and the data encryption in the 5G communication module adopts the AES-256 protocol; the variable air volume Venturi valve of the actuator is connected to the intelligent damper electric actuator through a Profibus-DP bus.
7. A method for controlling the gas environment of a clean laboratory based on the system according to any one of claims 1 to 6, characterized in that: The following steps are involved: Step 1: Multi-dimensional data collection and preprocessing: Spatial grid sampling: The orbital suction hood moves along a preset path, stops for 5 seconds at every 1m interval to collect gas samples, and generates a three-dimensional concentration distribution matrix in the laboratory; Data denoising: Kalman filtering algorithm is used to denoise sensor data, remove high-frequency noise, and pass threshold verification; Step 2: Intelligent analysis and strategy generation: Three-dimensional airflow simulation: Input the laboratory geometry model, which includes obstacles such as fume hoods and laboratory tables, the location of air supply and exhaust vents, and current temperature and humidity parameters, calculate the airflow field, and predict the time threshold for pollutants to diffuse into the buffer zone; Reinforcement learning optimization: construct state space S, action space A, and reward function R, and generate the optimal control strategy through Q-learning algorithm; the state space S includes temperature and humidity, pressure, and pollutant concentration; the action space A includes supply and exhaust air volume, filtration level, and temperature and humidity setting values; the reward function R = energy consumption × 0.1-∑|parameter deviation|×0.9; Step 3: Dynamic Control and Execution: Filtration system control: When PM2.5 concentration is greater than 10 μg / m 3 When the TVOC concentration is greater than 10ppm and the activated carbon adsorption module operation time is greater than 4 hours, the thermal regeneration program is triggered; Airflow and pressure control: Based on the CFD simulation results, the flow rate of each air supply and exhaust port is adjusted through the variable air volume Venturi valve to maintain the average wind speed in the clean area section at 0.3-0.5m / s; When the dynamic pressure sensor detects that the pressure difference between the clean area and the buffer area deviates from -35Pa±2Pa, the pressure compensation algorithm is started; Temperature and humidity control: Fuzzy PID control algorithm is used, with temperature setting value of 20-25℃ and humidity setting value of 40%-60%RH. The air supply temperature is dynamically adjusted according to the density of laboratory personnel. Step 4: Remote monitoring and system maintenance: Real-time warning mechanism: when the bioaerosol concentration is greater than 100 CFU / m 3 Or when the CO2 concentration is greater than 1500ppm, the cloud platform notifies the responsible person through sound and light alarms and text messages, and automatically generates a maintenance work order; Energy consumption management module: predict the energy consumption of the next day based on the ARIMA model, and optimize the compressor frequency and fresh air ratio of the VRF air-conditioning system.
8. The method for controlling the gas environment of a clean laboratory according to claim 7, characterized in that: The time series database updates the flow field data every 30 minutes, and automatically adjusts the air outlet angle when an eddy current area is detected.
Citation Information
Patent Citations
Shadow elimination light plate, shadow elimination table and shadowless lighting device
CN209801234U
Location awareness apparatus, vehicle having the same and method for controlling the apparatus
US10234567B2
Cited By
Construction method of PVDF polymerization post-treatment clean space
CN121383317A
