Intelligent air volume and air pressure control system

Abstract

The invention discloses an air volume and air pressure intelligent control system, relates to the technical field of industrial automation control, and discloses the air volume and air pressure intelligent control system, which comprises six core modules: a data acquisition and preprocessing module acquires and preprocesses data through each sensor; the parameter threshold value setting and marking module sets a conventional threshold value range, a comfort control threshold value and a safety control threshold value of parameters, and marks abnormal data exceeding the threshold values; the multi-parameter modeling and mode decision module uses a random forest regression algorithm to carry out modeling according to relevance, evaluates the model through cross validation, and selects a control mode according to data and a threshold value; the control mode execution module controls processing and execution of the three modes; the machine learning optimization module monitors data changes, dynamically adjusts model parameters, and optimizes air volume and air pressure adjustment based on a support vector regression algorithm; and the feedback optimization module updates the control model based on the feedback optimization control parameters.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial automation control technology, and in particular to an intelligent control system for air volume and air pressure. Background Art

[0002] With the rapid development of intelligent technology and industrial automation control technology, industrial environmental control systems are gradually developing towards more efficient, precise and adaptive directions. However, many traditional air volume and pressure control systems still face many shortcomings.

[0003] Existing air volume and pressure control systems rely on linear or simplified mathematical models and are unable to effectively capture the complex nonlinear relationship between air volume, air pressure and environmental parameters. The mode selection of traditional control systems is usually based on a simple switching mechanism, which can only switch between preset control modes and cannot perform flexible and intelligent control according to actual environmental changes. Existing technologies fail to comprehensively consider the impact of pollutants on the environment and human health, and lack a mechanism to sort the elimination priorities according to the severity of the impact and eliminate them in sequence, resulting in less than ideal air quality improvement effects. During the data monitoring process, existing systems can usually only monitor parameter changes in real time and lack the ability to adjust control strategies based on data feedback. The air volume and air pressure regulation of most control systems is based on simple control algorithms, and they fail to fully utilize advanced algorithms such as machine learning for dynamic optimization, and are unable to optimize the control process based on historical data and environmental changes. Summary of the Invention

[0004] The purpose of the present invention is to provide an intelligent control system for air volume and air pressure to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] In a first aspect, the present invention provides an intelligent air volume and pressure control system, comprising:

[0007] Data acquisition and preprocessing module: The data acquisition unit collects temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration through various sensors; the data preprocessing unit cleans, standardizes and normalizes the data;

[0008] Parameter threshold setting and marking module: The threshold setting unit sets the normal threshold range, comfort control threshold and safety control threshold of the parameter; the parameter marking unit marks abnormal data that exceeds the threshold;

[0009] Multi-parameter modeling and mode decision module: The model design and training unit uses the random forest regression algorithm to build models based on the correlation between wind volume and wind pressure and other parameters. The model is trained based on historical data and evaluated through cross-validation. The mode selection unit selects the control mode based on data and thresholds.

[0010] Control mode execution module: The conventional control mode unit adjusts air volume and pressure to ensure that other parameters are within the normal threshold range; the comfort control mode unit adjusts air volume and pressure to adjust temperature and humidity; the safety control mode unit calculates the severity of the impact of pollutants on the environment and human health, sorts the elimination priorities based on the final severity of the impact, and then eliminates them in sequence, then adjusts air volume and pressure to expel the remaining pollutants;

[0011] Machine learning optimization module: The data monitoring and analysis unit monitors and analyzes data changes in real time; the model parameter dynamic adjustment unit adjusts model parameters based on feedback; and the air volume and pressure control model unit optimizes air volume and pressure regulation based on the support vector regression algorithm.

[0012] Feedback optimization module: Among them, the parameter optimization unit optimizes the parameters in the control algorithm based on feedback; the model update unit regularly updates the control model.

[0013] In conjunction with the first aspect, in a first implementation of the first aspect of the present application, the data acquisition unit collects temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration through various sensors; and the data preprocessing unit cleans, standardizes, and normalizes the data, including:

[0014] The data acquisition unit collects the temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration in the factory through sensors, including temperature sensors, humidity sensors, CO2 concentration sensors, PM2.5 concentration sensors and VOCs concentration sensors; the data preprocessing unit removes noise, detects and fills the collected data, converts the collected data into a unified standard range, and converts the data value into the [0, 1] interval.

[0015] In combination with the first aspect, in a second implementation of the first aspect of the present application, the threshold setting unit sets a normal threshold range, a comfort control threshold, and a safety control threshold of the parameter; and the parameter marking unit marks abnormal data that exceeds the threshold, including:

[0016] The threshold setting unit sets the normal threshold range, comfort control threshold and safety control threshold for parameters including temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration, and defines them according to the equipment operation standards, environmental standards and the threshold range set by the user; during the data collection process, each collected data is compared with the preset threshold to determine whether it exceeds the threshold; when the collected data exceeds the threshold, the parameter marking unit automatically marks the data as abnormal data, records it and notifies the corresponding responsible personnel.

[0017] In combination with the first aspect, in a third implementation of the first aspect of the present application, the model design and training unit uses a random forest regression algorithm to model according to the correlation between wind volume and wind pressure and other parameters, trains based on historical data, and evaluates the model through cross-validation; the mode selection unit selects a control mode based on the data and threshold, including:

[0018] The model design and training unit selects key features related to wind volume and pressure prediction based on data relevance and actual needs, including temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration. Using the random forest regression algorithm, the preprocessed dataset is divided into training and test sets. Multiple decision trees are generated. During the training process of each tree, a portion of training samples is randomly selected, and features are randomly selected at each node for splitting. Each tree is trained on the randomly selected sample subset and feature subset, and the prediction results of all trees are averaged to obtain the predicted value.

[0019] Through K-fold cross-validation, the dataset is divided into K subsets, and K training and testing are performed. Each time, K-1 subsets are used as training sets, and the remaining subset is used as the test set.

[0020] The training model is used to predict the test set, and the mean square error of each test is calculated to measure the difference between the model prediction value and the actual value. The smaller the value, the more accurate the model is. The formula is:

[0021]

[0022] Among them, MSE is the mean square error, y i is the actual value, is the predicted value, and N is the number of samples in the test set.

[0023] In combination with the first aspect, in a fourth implementation of the first aspect of the present application, the mode selection unit selects a control mode according to the data and the threshold, including:

[0024] The mode selection unit compares the real-time monitoring data with the preset threshold value to determine and switch the control mode;

[0025] When all parameters are within the preset normal threshold range, the normal control mode is automatically selected; when the temperature or humidity exceeds the preset comfort control threshold, it switches to the comfort control mode; when the CO2 concentration, PM2.5 concentration or VOCs concentration exceeds the preset safety control threshold, it switches to the safety control mode to treat the polluted gas.

[0026] In combination with the first aspect, in a fifth implementation of the first aspect of the present application, the conventional control mode unit adjusts the air volume and air pressure to ensure that other parameters are within a conventional threshold range, including:

[0027] Based on real-time monitoring data, the conventional control mode unit adjusts the air volume to ensure that air circulation meets the preset standards, so that the temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration parameters are kept within the conventional threshold range, preventing these parameters from exceeding the set range and ensuring the stability of the working environment;

[0028] The conventional control mode unit adjusts the air pressure according to the air flow and air quality requirements in the factory to ensure uniform air circulation in the factory. It continuously monitors temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration data, and dynamically adjusts the air volume and pressure to ensure that all parameters are within the preset conventional threshold range. It uses the PID control algorithm to adjust the air volume and pressure by performing proportional, integral, and differential processing on the error. The formula is:

[0029]

[0030] Among them, u(t) is the controlled variable of air volume or air pressure, e(t) is the error, t is the actual control time, τ is the time variable of integration, e(τ) is the error function, d is the differential operator, dt is the small time increment in the differential process, K p , K i , K d are the proportional, integral and differential coefficients respectively.

[0031] In combination with the first aspect, in a sixth implementation of the first aspect of the present application, the comfort control mode unit adjusts the temperature and humidity by adjusting the air volume and air pressure, including:

[0032] Based on the monitored temperature and humidity data, the comfort control mode unit adjusts the air volume, adjusts the indoor air circulation speed, and adjusts the indoor temperature and humidity to maintain them within the preset comfort control thresholds; by adjusting the wind pressure, it ensures the uniformity and efficiency of air flow and optimizes the air distribution, so that the temperature and humidity are evenly adjusted throughout the space, avoiding local adjustment;

[0033] Based on the real-time data feedback from the temperature sensor and humidity sensor, the air volume and pressure are dynamically adjusted to respond to changes in temperature and humidity in real time.

[0034] In combination with the first aspect, in a seventh implementation of the first aspect of the present application, the safety control mode unit calculates the severity of the impact of the pollutant gas on the environment and human health, sorts the elimination priorities according to the final impact severity, performs elimination in sequence, and then adjusts the air volume and air pressure to discharge the residual pollutant gas, including:

[0035] The safety control mode unit monitors CO2 concentration, PM2.5 concentration, and VOCs concentration in real time, assesses the severity of each pollutant's impact on the environment and human health, assigns a weight coefficient, and calculates the impact of pollutants on the environment using the following formulas:

[0036] Iep=CCp×ep;

[0037] Iec=Cc×ec;

[0038] Ieυ=Cυ×eυ;

[0039] Among them, I ep , I ec and I ev The impacts of PM2.5, CO2 and VOCs on the environment, C p 、C c and C v are the concentrations of PM2.5, CO2 and VOCs, respectively, e p 、e c and e v are the weight coefficients of the environmental impacts of PM2.5, CO2, and VOCs;

[0040] The formulas for calculating the impact of pollutant gases on human health are:

[0041] Ihp=Cp×hp;

[0042] Ihc=Cc×hc;

[0043] Ihv=Cυ×hυ;

[0044] Among them, I hp , I hc and I hv The impacts of PM2.5, CO2 and VOCs on human health, C p 、C c and C v are the concentrations of PM2.5, CO2 and VOCs, respectively, h p 、h c and hv are the weight coefficients of the impact of PM2.5, CO2 and VOCs on human health;

[0045] According to the impact of pollutant gases on the environment and human health, the final severity of the impact is calculated using the following formulas:

[0046] Fp=α×Iep+(1-α)×Ihp;

[0047] Fc = α × Iec + (1-α) × Ihc;

[0048] Fv=α×Iev+(1-α)×Ihv;

[0049] Among them, F p 、F c and F v are the final impact severity of PM2.5, CO2 and VOCs, respectively; α is the weight coefficient of environmental impact on the final impact severity; (1-α) is the weight coefficient of human health impact on the final impact severity;

[0050] Compare the severity of the final impact of pollutant gases and determine the priority of pollutant gas removal;

[0051] A water spray tower is used to absorb and remove PM2.5 particles from the air, reducing the concentration of particulate matter in the air; a palladium catalyst is used to promote the oxidation reaction of VOCs, converting them into harmless carbon dioxide and water, reducing the impact of toxic gases on the environment; a sodium hydroxide solution is used to absorb CO2, and through a chemical reaction, CO2 is converted into sodium carbonate, reducing the concentration of carbon dioxide in the air; after the polluted gas is digested, the safety control mode unit adjusts the air volume and pressure according to the changes in the concentration of the polluted gas to ensure unobstructed air circulation and discharge the remaining polluted gas.

[0052] In combination with the first aspect, in an eighth implementation of the first aspect of the present application, the data monitoring and analysis unit monitors data changes in real time and performs analysis; the model parameter dynamic adjustment unit adjusts model parameters based on feedback; and the air volume and air pressure control model unit optimizes air volume and air pressure adjustment based on a support vector regression algorithm, including:

[0053] The data monitoring and analysis unit monitors changes in temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration in the factory in real time, performs dynamic analysis, collects new sensor data, and calculates the current environmental conditions. The model parameter dynamic adjustment unit dynamically adjusts the parameters in the air volume and air pressure control model based on feedback from real-time monitoring data.

[0054] The air volume and pressure control model unit is based on the support vector regression algorithm. Through training on historical data, it establishes a nonlinear relationship between air volume, air pressure and environmental parameters, and adjusts the air volume and air pressure in response to changes in temperature, humidity and pollutant gas concentration in the factory.

[0055] In conjunction with the first aspect, in a ninth implementation of the first aspect of the present application, the parameter optimization unit optimizes the parameters in the control algorithm based on feedback; and the model updating unit periodically updates the control model, including:

[0056] The parameter optimization unit optimizes the parameters in the control algorithm based on real-time feedback data, including the PID control algorithm and the support vector regression algorithm, analyzes the impact of the control parameters on the air volume and pressure regulation, and adjusts the parameters through the genetic algorithm;

[0057] The model update unit regularly updates the control model to ensure that the model can adapt to changes in the factory environment and new operational requirements, retrains the existing model using new monitoring data, and optimizes model performance by introducing new data features, adjusting the model structure, and updating training parameters.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] 1. The present invention uses a random forest regression algorithm to capture the nonlinear relationship between wind volume and pressure and other parameters, models them based on correlation, trains them based on historical data, and evaluates the model through cross-validation.

[0060] 2. The present invention selects conventional control mode, comfort control mode and safety control mode units based on the collected data and preset thresholds; among them, the safety control mode is calculated based on the severity of the impact of pollutant gases on the environment and human health, and the elimination priority is sorted by the final severity of the impact, and the elimination is carried out in sequence, and then the air volume and air pressure are adjusted to discharge the residual pollutant gas.

[0061] 3. The present invention monitors data changes in real time and analyzes them, optimizes and adjusts the air volume and air pressure in combination with the support vector regression algorithm, adjusts model parameters according to feedback, and optimizes the air volume and air pressure regulation based on the support vector regression algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a system structure diagram of an intelligent control system for air volume and pressure according to the present invention;

[0063] Figure 2 This is a schematic diagram of the control mode execution module steps of an intelligent air volume and pressure control system of the present invention. DETAILED DESCRIPTION

[0064] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0065] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution.

[0066] like Figure 1 As shown in the system structure diagram of an intelligent control system for air volume and pressure, the present invention provides an intelligent control system for air volume and pressure, comprising:

[0067] Step S100: The data acquisition unit collects temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration through various sensors; the data preprocessing unit cleans, standardizes and normalizes the data;

[0068] Specifically, the data acquisition unit collects the temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration in the factory through sensors, including temperature sensors, humidity sensors, CO2 concentration sensors, PM2.5 concentration sensors and VOCs concentration sensors; the data preprocessing unit removes noise, detects and fills the collected data, converts the collected data into a unified standard range, and converts the data value into the [0, 1] interval.

[0069] In a specific embodiment, an industrial plant in Shanghai is selected as the collection location, and environmental data is collected in real time during a certain period of time. The average temperature is 22.5°C, the average humidity is 55%, the average CO2 concentration is 400ppm, and the average PM2.5 concentration is 30μg / m 3 , the average VOCs concentration is 150ppb.

[0070] The collected raw data is cleaned, standardized and normalized. The PM2.5 concentration sensor collects data of 1000μg / m 3 , the data is determined to be noise and is removed. Standardization converts the data to a standard normal distribution with zero mean and unit variance, and normalization converts the data to the interval [0, 1]. The normalized values ​​of temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration are 0.5, 0.5, 0.1429, 0.06, and 0.3, respectively.

[0071] Step S200: The threshold setting unit sets the normal threshold range, the comfort control threshold, and the safety control threshold of the parameter; the parameter marking unit marks abnormal data that exceeds the threshold;

[0072] Specifically, the threshold setting unit sets the normal threshold range, comfort control threshold and safety control threshold for parameters including temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration, and defines them according to the equipment operation standards, environmental standards and the threshold range set by the user; during the data collection process, each collected data is compared with the preset threshold to determine whether it exceeds the threshold; when the collected data exceeds the threshold, the parameter marking unit automatically marks the data as abnormal data, records it and notifies the corresponding responsible personnel.

[0073] In a specific embodiment, according to the equipment operation standard, environmental standard and the threshold range set by the user, a specific conventional threshold range, a comfort control threshold and a safety control threshold are set for the parameter. For temperature, the conventional threshold range is set to [18°C, 26°C], and the comfort control threshold is set to [20°C, 24°C]; for humidity, the conventional threshold range is set to [40%, 60%], and the comfort control threshold is set to [45%, 55%]; for CO2 concentration, the conventional threshold range is set to [350ppm, 800ppm], and the safety control threshold is set to [300ppm, 1000ppm]; for PM2.5 concentration, the conventional threshold range is set to [0μg / m 3 , 50 μg / m 3 ], the safety control threshold is set at [0μg / m 3 , 100 μg / m 3 ]; for VOCs concentration, the conventional threshold range is set to [0ppb, 200ppb], and the safety control threshold is set to [0ppb, 500ppb].

[0074] The collected data was compared with the threshold values, where the temperature was 27°C, the humidity was 55%, the CO2 concentration was 450ppm, and the PM2.5 concentration was 80μg / m 3 , the VOCs concentration is 150ppb, where the temperature exceeds the normal threshold range, it is marked as abnormal data, an abnormal report is generated, and the corresponding responsible personnel are notified.

[0075] Step S300: The model design and training unit uses a random forest regression algorithm to build a model based on the correlation between wind volume and wind pressure and other parameters, trains based on historical data, and evaluates the model through cross-validation; the mode selection unit selects a control mode based on the data and threshold;

[0076] Specifically, the model design and training unit selects key features related to wind volume and pressure prediction based on data relevance and actual needs, including temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration. Using the random forest regression algorithm, the preprocessed dataset is divided into a training set and a test set. Multiple decision trees are generated. During the training process of each tree, a portion of training samples is randomly selected, and features are randomly selected at each node for splitting. Each tree is trained on the randomly selected sample subset and feature subset, and the prediction results of all trees are averaged to obtain the predicted value.

[0077] Through K-fold cross-validation, the dataset is divided into K subsets, and K training and testing are performed. Each time, K-1 subsets are used as training sets, and the remaining subset is used as the test set.

[0078] The training model is used to predict the test set, and the mean square error of each test is calculated to measure the difference between the model prediction value and the actual value. The smaller the value, the more accurate the model is. The formula is:

[0079]

[0080] Among them, MSE is the mean square error, y i is the actual value, is the predicted value, and N is the number of samples in the test set.

[0081] Furthermore, the mode selection unit compares the real-time monitoring data with a preset threshold value to determine and switch the control mode;

[0082] When all parameters are within the preset normal threshold range, the normal control mode is automatically selected; when the temperature or humidity exceeds the preset comfort control threshold, it switches to the comfort control mode; when the CO2 concentration, PM2.5 concentration or VOCs concentration exceeds the preset safety control threshold, it switches to the safety control mode to treat the polluted gas.

[0083] In one specific embodiment, a random forest regression algorithm was used to construct a prediction model by analyzing the correlation between temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration in historical data and air volume and pressure. During model training, a dataset containing 10,000 records was divided into a training set (80% of the data) and a test set (20% of the data) with 8,000 records. 100 decision trees were generated, with each tree randomly splitting on 70% of the samples and 5 features during training. Using 5-fold cross-validation, the calculated mean square error was 15, indicating that the model has high prediction accuracy.

[0084] Mode switching is performed based on the comparison between real-time data and threshold values. If the current temperature is 23°C, humidity is 48%, CO2 concentration is 450ppm, and PM2.5 concentration is 38μg / m 3 , VOCs concentration is 50ppb, both within the normal threshold range, and the normal control mode is automatically selected. If the monitored PM2.5 concentration rises suddenly to 110μg / m 3 , exceeding the safety threshold [0μg / m 3 , 100 μg / m 3 ], and automatically switches to safety control mode immediately.

[0085] Step S400: The normal control mode unit adjusts the air volume and pressure to ensure that other parameters are within the normal threshold range. The comfort control mode unit adjusts the air volume and pressure to adjust the temperature and humidity. The safety control mode unit calculates the severity of the impact of the pollutants on the environment and human health, sorts the elimination priorities based on the final severity of the impact, and then eliminates them in sequence, then adjusts the air volume and pressure to expel the remaining pollutants.

[0086] Specifically, based on real-time monitoring data, the conventional control mode unit adjusts the air volume to ensure that air circulation reaches the preset standard, so that the temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration parameters are kept within the conventional threshold range, preventing these parameters from exceeding the set range and ensuring the stability of the working environment;

[0087] The conventional control mode unit adjusts the air pressure according to the air flow and air quality requirements in the factory to ensure uniform air circulation in the factory. It continuously monitors temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration data, and dynamically adjusts the air volume and pressure to ensure that all parameters are within the preset conventional threshold range. It uses the PID control algorithm to adjust the air volume and pressure by performing proportional, integral, and differential processing on the error. The formula is:

[0088]

[0089] Among them, u(t) is the controlled variable of air volume or air pressure, e(t) is the error, t is the actual control time, τ is the time variable of integration, e(τ) is the error function, d is the differential operator, dt is the small time increment in the differential process, K p , K i , K d are the proportional, integral and differential coefficients respectively.

[0090] Based on the monitored temperature and humidity data, the comfort control mode unit adjusts the air volume, adjusts the indoor air circulation speed, and adjusts the indoor temperature and humidity to maintain them within the preset comfort control thresholds; by adjusting the wind pressure, it ensures the uniformity and efficiency of air flow and optimizes the air distribution, so that the temperature and humidity are evenly adjusted throughout the space, avoiding local adjustment;

[0091] Based on the real-time data feedback from the temperature sensor and humidity sensor, the air volume and pressure are dynamically adjusted to respond to changes in temperature and humidity in real time.

[0092] The safety control mode unit monitors CO2 concentration, PM2.5 concentration, and VOCs concentration in real time, assesses the severity of each pollutant's impact on the environment and human health, assigns a weight coefficient, and calculates the impact of pollutants on the environment using the following formulas:

[0093] Iep=CCp×ep;

[0094] Iec=Cc×ec;

[0095] Ieυ=Cυ×eυ;

[0096] Among them, I ep , I ec and I ev The impacts of PM2.5, CO2 and VOCs on the environment, C p 、C c and C v are the concentrations of PM2.5, CO2 and VOCs, respectively, e p 、e c and e v are the weight coefficients of the environmental impacts of PM2.5, CO2, and VOCs;

[0097] The formulas for calculating the impact of pollutant gases on human health are:

[0098] Ihp=Cp×hp;

[0099] Ihc=Cc×hc;

[0100] Ihυ=Cυ×hυ;

[0101] Among them, I hp , I hc and I hv The impacts of PM2.5, CO2 and VOCs on human health, C p 、C c and C v are the concentrations of PM2.5, CO2 and VOCs, respectively, h p 、h c and hv are the weight coefficients of the impact of PM2.5, CO2 and VOCs on human health;

[0102] According to the impact of pollutant gases on the environment and human health, the final severity of the impact is calculated using the following formulas:

[0103] Fp=α×Iep+(1-α)×Ihp;

[0104] Fc = α × Iec + (1-α) × Ihc;

[0105] Fv=α×Iev+(1-α)×Ihv;

[0106] Among them, F p 、F c and F v are the final impact severity of PM2.5, CO2 and VOCs, respectively; α is the weight coefficient of environmental impact on the final impact severity; (1-α) is the weight coefficient of human health impact on the final impact severity;

[0107] Compare the severity of the final impact of pollutant gases and determine the priority of pollutant gas removal;

[0108] A water spray tower is used to absorb and remove PM2.5 particles from the air, reducing the concentration of particulate matter in the air; a palladium catalyst is used to promote the oxidation reaction of VOCs, converting them into harmless carbon dioxide and water, reducing the impact of toxic gases on the environment; a sodium hydroxide solution is used to absorb CO2, and through a chemical reaction, CO2 is converted into sodium carbonate, reducing the concentration of carbon dioxide in the air; after the polluted gas is digested, the safety control mode unit adjusts the air volume and pressure according to the changes in the concentration of the polluted gas to ensure unobstructed air circulation and discharge the remaining polluted gas.

[0109] In a specific embodiment, the measured data are all within the conventional threshold range, the target temperature is set to the center of the conventional threshold range, that is, 22°C, the measured temperature is 23.5°C, the error e(t) is -1.5°C, and the proportional coefficient K is set. p , integral coefficient K i and differential coefficient K d are 5, 0.1 and 2 respectively, according to the formula:

[0110]

[0111] The calculation shows that u(t)=-7.5, and the air volume and air pressure are adjusted in the opposite direction.

[0112] When the PM2.5 concentration, CO2 concentration and VOCs concentration were measured at a certain time period, they were 70 μg / m 3When the concentration of pollutants reaches 1200ppm and 300ppb, the system compares the concentration with the threshold value and switches to the safety control mode. The impact of pollutants on the environment is calculated using the following formula:

[0113] Iep=CCp×ep;

[0114] Iec=Cc×ec;

[0115] Iev=CCv×ev;

[0116] C p =70, C c =1200 and C v =300,e p 、e c and e v Set them to 0.3, 0.2 and 0.5 respectively, and calculate I ep , I ec and I ev 21, 240 and 150 respectively;

[0117] The impact of pollutant gases on human health is calculated using the formula:

[0118] Ihp=Cp×hp;

[0119] Ihc=Cc×hc;

[0120] Ihv=Cυ×hv;

[0121] Set h p 、h c and h v are 0.6, 0.8 and 0.4 respectively, and the calculated I hp , I hc and I hv They are 42, 960 and 120 respectively;

[0122] According to the impact of pollutant gases on the environment and human health, the final severity of the impact is calculated using the formula:

[0123] Fp=α×Iep+(1-α)×Ihp;

[0124] Fc = α × Iec + (1-α) × Ihc;

[0125] Fv=α×Iev+(1-α)×Ihv;

[0126] Set α to 0.6 and (1-α) to 0.4, and calculate F p 、F c and F v 29.4, 528 and 138 respectively;

[0127] By comparing the severity of the final impact of pollutant gases, it is concluded that the priority order of pollutant gas removal is CO2, VOCs, and PM2.5.

[0128] CO2 is treated first, and sodium hydroxide solution is used to absorb CO2, and CO2 is converted into sodium carbonate through a chemical reaction; then palladium catalyst is used to treat VOCs, promoting the oxidation reaction of VOCs and converting them into harmless carbon dioxide and water. Finally, a water spray tower is used to adsorb and remove PM2.5 particles in the air. After the polluted gas is digested, the air volume and pressure are adjusted according to the changes in the concentration of the polluted gas to ensure smooth air circulation and discharge the remaining polluted gas.

[0129] Step S500: The data monitoring and analysis unit monitors data changes in real time and performs analysis; the model parameter dynamic adjustment unit adjusts the model parameters according to feedback; the air volume and air pressure control model unit optimizes the air volume and air pressure adjustment based on the support vector regression algorithm; the parameter optimization unit optimizes the parameters in the control algorithm based on feedback; and the model update unit regularly updates the control model.

[0130] Specifically, the data monitoring and analysis unit monitors the changes in temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration in the factory in real time, performs dynamic analysis, collects new sensor data, and calculates the current environmental conditions. The model parameter dynamic adjustment unit dynamically adjusts the parameters in the air volume and air pressure control model based on feedback from real-time monitoring data.

[0131] The air volume and pressure control model unit is based on the support vector regression algorithm. Through training on historical data, it establishes a nonlinear relationship between air volume, air pressure and environmental parameters, and adjusts the air volume and air pressure in response to changes in temperature, humidity and pollutant gas concentration in the factory.

[0132] Furthermore, the parameter optimization unit optimizes the parameters in the control algorithm based on real-time feedback data, including the PID control algorithm and the support vector regression algorithm, analyzes the impact of the control parameters on the air volume and pressure regulation, and adjusts the parameters through the genetic algorithm;

[0133] The model update unit regularly updates the control model to ensure that the model can adapt to changes in the factory environment and new operational requirements, retrains the existing model using new monitoring data, and optimizes model performance by introducing new data features, adjusting the model structure, and updating training parameters.

[0134] In a specific embodiment, various environmental parameters in the factory are collected in real time and continuously analyzed to determine whether the current environment is within the preset threshold. When the CO2 concentration is monitored to rise to 900ppm, the support vector regression model predicts that the current air volume needs to be increased from 500m 3 / h adjusted to 550m3 / h, and the wind pressure was adjusted from 150Pa to 155Pa to more effectively dilute the indoor air and reduce CO2 concentrations. Using a genetic algorithm, based on the last 24 hours of real-time feedback data, the penalty coefficient C of the PID control algorithm and support vector regression model was optimized from 5.0 to 4.8 to improve the model's prediction accuracy and robustness under current operating conditions.

[0135] The model update unit is set to execute every 14 days, retraining the existing support vector regression control model using all new monitoring data accumulated in the past 14 days. The existing model structure is evaluated to determine whether the kernel function type needs to be changed. The model is then retrained using the updated dataset to adapt to seasonal environmental changes or new production operation requirements that may occur within the plant.

[0136] like Figure 2 As shown in the schematic diagram of the control mode execution module steps of an intelligent control system for air volume and air pressure, the present invention provides an intelligent control system for air volume and air pressure, comprising:

[0137] The three mode units of conventional control, comfort control, and safety control work together, and are processed according to the control requirements of different mode units based on real-time monitoring data, preset threshold ranges, and dynamic models. The specific steps are as follows:

[0138] Real-time monitoring data, including temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration, are input. Different control modes are triggered by the following conditions during execution: when all parameters are within the normal threshold range, the normal control mode is automatically selected; when the temperature or humidity exceeds the comfort control threshold, the control mode is switched to the comfort control mode; when the CO2 concentration, PM2.5 concentration, or VOCs concentration exceeds the safety control threshold, the control mode is switched to the safety control mode.

[0139] The conventional control mode unit adjusts air volume based on real-time monitoring data to ensure that air circulation meets preset standards, keeping temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration within conventional thresholds. Air pressure is adjusted based on airflow and air quality requirements within the factory to ensure even air circulation within the factory. Temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration data are continuously monitored, and air volume and pressure are dynamically adjusted to ensure that all parameters remain within the preset conventional thresholds. A PID control algorithm is used to adjust air volume and pressure by performing proportional, integral, and differential processing on errors.

[0140] The comfort control mode unit adjusts air volume and indoor air circulation speed based on monitored temperature and humidity data to maintain indoor temperature and humidity within preset comfort control thresholds. It also adjusts air pressure to ensure uniformity and efficiency of air flow, optimizing air distribution and ensuring uniform temperature and humidity throughout the space, avoiding localized adjustments. Air volume and pressure are dynamically adjusted based on real-time data feedback from temperature and humidity sensors, responding to changes in temperature and humidity in real time.

[0141] The safety control mode unit monitors CO2, PM2.5, and VOC concentrations in real time, assessing the severity of each pollutant's impact on the environment and human health. It then assigns a weighting factor and uses this weighting factor to calculate the impact of each pollutant on the environment and human health. Formulas are used to calculate the impact of each pollutant on the environment and human health, respectively. The final severity of the impact is then calculated based on the impacts of each pollutant on the environment and human health. The final severity of the impacts is then compared to determine the priority for pollutant removal.

[0142] Polluted gases undergo different treatments, using a water spray tower to absorb and remove PM2.5 particles from the air, reducing the concentration of particulate matter in the air. A palladium catalyst promotes the oxidation of VOCs, converting them into harmless carbon dioxide and water, reducing the environmental impact of toxic gases. Sodium hydroxide solution absorbs CO2 and chemically converts it into sodium carbonate, reducing the concentration of carbon dioxide in the air. After the polluted gases are digested, the safety control mode unit adjusts the air volume and pressure based on changes in polluted gas concentration to ensure unobstructed air circulation and expel any remaining polluted gases.

[0143] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. An intelligent control system for air volume and pressure, characterized in that: include: Data acquisition and preprocessing module: The data acquisition unit collects temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration through various sensors; the data preprocessing unit cleans, standardizes and normalizes the data; Parameter threshold setting and marking module: The threshold setting unit sets the normal threshold range, comfort control threshold and safety control threshold of the parameter; the parameter marking unit marks abnormal data that exceeds the threshold; Multi-parameter modeling and mode decision module: The model design and training unit uses the random forest regression algorithm to build models based on the correlation between wind volume and wind pressure and other parameters. The model is trained based on historical data and evaluated through cross-validation. The mode selection unit selects the control mode based on data and thresholds. Control mode execution module: The conventional control mode unit adjusts air volume and pressure to ensure that other parameters are within the normal threshold range; the comfort control mode unit adjusts air volume and pressure to adjust temperature and humidity; the safety control mode unit calculates the severity of the impact of pollutants on the environment and human health, sorts the elimination priorities based on the final severity of the impact, and then eliminates them in sequence, then adjusts air volume and pressure to expel the remaining pollutants; Machine learning optimization module: The data monitoring and analysis unit monitors and analyzes data changes in real time; the model parameter dynamic adjustment unit adjusts model parameters based on feedback; and the air volume and pressure control model unit optimizes air volume and pressure regulation based on the support vector regression algorithm. Feedback optimization module: Among them, the parameter optimization unit optimizes the parameters in the control algorithm based on feedback; the model update unit regularly updates the control model.

2. The intelligent control system for air volume and pressure according to claim 1, characterized in that: The data acquisition unit collects temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration through various sensors; The data preprocessing unit cleans, standardizes and normalizes the data, including: The data acquisition unit collects the temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration in the factory through sensors, including temperature sensors, humidity sensors, CO2 concentration sensors, PM2.5 concentration sensors and VOCs concentration sensors; the data preprocessing unit removes noise, detects and fills the collected data, converts the collected data into a unified standard range, and converts the data value into the [0, 1] interval.

3. The intelligent control system for air volume and pressure according to claim 1, characterized in that: The threshold setting unit sets a conventional threshold range, a comfort control threshold, and a safety control threshold of a parameter; The parameter marking unit marks abnormal data that exceeds the threshold, including: The threshold setting unit sets the normal threshold range, comfort control threshold and safety control threshold for parameters including temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration, and defines them according to the equipment operation standards, environmental standards and the threshold range set by the user; during the data collection process, each collected data is compared with the preset threshold to determine whether it exceeds the threshold; when the collected data exceeds the threshold, the parameter marking unit automatically marks the data as abnormal data, records it and notifies the corresponding responsible personnel.

4. The intelligent control system for air volume and pressure according to claim 1, characterized in that: The model design and training unit uses a random forest regression algorithm to build a model based on the correlation between wind volume and wind pressure and other parameters, trains based on historical data, and evaluates the model through cross-validation; The mode selection unit selects the control mode according to the data and threshold, including: The model design and training unit selects key features related to wind volume and pressure prediction based on data relevance and actual needs, including temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration. Using the random forest regression algorithm, the preprocessed dataset is divided into training and test sets. Multiple decision trees are generated. During the training process of each tree, a portion of training samples is randomly selected, and features are randomly selected at each node for splitting. Each tree is trained on the randomly selected sample subset and feature subset, and the prediction results of all trees are averaged to obtain the predicted value. Through K-fold cross-validation, the dataset is divided into K subsets, and K training and testing are performed. Each time, K-1 subsets are used as training sets, and the remaining subset is used as the test set. The training model is used to predict the test set, and the mean square error of each test is calculated to measure the difference between the model prediction value and the actual value. The smaller the value, the more accurate the model is. The formula is: Among them, MSE is the mean square error, y i is the actual value, is the predicted value, and N is the number of samples in the test set.

5. The intelligent control system for air volume and pressure according to claim 1, characterized in that: The mode selection unit selects a control mode according to the data and the threshold, including: The mode selection unit compares the real-time monitoring data with the preset threshold value to determine and switch the control mode; When all parameters are within the preset normal threshold range, the normal control mode is automatically selected; when the temperature or humidity exceeds the preset comfort control threshold, it switches to the comfort control mode; when the CO2 concentration, PM2.5 concentration or VOCs concentration exceeds the preset safety control threshold, it switches to the safety control mode to treat the polluted gas.

6. The intelligent control system for air volume and pressure according to claim 1, characterized in that: The conventional control mode unit adjusts the air volume and pressure to ensure that other parameters are within the conventional threshold range, including: Based on real-time monitoring data, the conventional control mode unit adjusts the air volume to ensure that air circulation meets the preset standards, so that the temperature, humidity, CO2 concentration, PM2.5 concentration and VOCs concentration parameters are kept within the conventional threshold range, preventing these parameters from exceeding the set range and ensuring the stability of the working environment; The conventional control mode unit adjusts the air pressure according to the air flow and air quality requirements in the factory to ensure uniform air circulation in the factory. It continuously monitors temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration data, and dynamically adjusts the air volume and pressure to ensure that all parameters are within the preset conventional threshold range. It uses the PID control algorithm to adjust the air volume and pressure by performing proportional, integral, and differential processing on the error. The formula is: Among them, u(t) is the controlled variable of air volume or air pressure, e(t) is the error, t is the actual control time, τ is the time variable of integration, e(τ) is the error function, d is the differential operator, dt is the small time increment in the differential process, K p , K i , K d are the proportional, integral and differential coefficients respectively.

7. The intelligent control system for air volume and pressure according to claim 1, characterized in that: The comfort control mode unit adjusts the temperature and humidity by adjusting the air volume and air pressure, including: Based on the monitored temperature and humidity data, the comfort control mode unit adjusts the air volume, adjusts the indoor air circulation speed, and adjusts the indoor temperature and humidity to maintain them within the preset comfort control thresholds; by adjusting the wind pressure, it ensures the uniformity and efficiency of air flow and optimizes the air distribution, so that the temperature and humidity are evenly adjusted throughout the space, avoiding local adjustment; Based on the real-time data feedback from the temperature sensor and humidity sensor, the air volume and pressure are dynamically adjusted to respond to changes in temperature and humidity in real time.

8. The intelligent control system for air volume and pressure according to claim 1, characterized in that: The safety control mode unit calculates the severity of the impact of pollutant gases on the environment and human health, sorts the elimination priorities according to the final severity of the impact, performs elimination in sequence, and then adjusts the air volume and air pressure to discharge the residual pollutant gases, including: The safety control mode unit monitors CO2 concentration, PM2.5 concentration, and VOCs concentration in real time, assesses the severity of each pollutant's impact on the environment and human health, assigns a weight coefficient, and calculates the impact of pollutants on the environment using the following formulas: Iep=Cp×ep; Ie=Cc×ec; Iev=Cv×ev; Among them, I ep , I ec and I ev The impacts of PM2.5, CO2 and VOCs on the environment, C p 、C c and C v are the concentrations of PM2.5, CO2 and VOCs, respectively, e p 、e c and e v are the weight coefficients of the environmental impacts of PM2.5, CO2, and VOCs; The formulas for calculating the impact of pollutant gases on human health are: Ihp=Cp×hp; Ihc=Cc×hc; Ihv=Cv×hv; Among them, I hp , I hc and I hv The impacts of PM2.5, CO2 and VOCs on human health, C p 、C c and C v are the concentrations of PM2.5, CO2 and VOCs, respectively, h p 、h c and h v are the weight coefficients of the impact of PM2.5, CO2 and VOCs on human health; According to the impact of pollutant gases on the environment and human health, the final severity of the impact is calculated using the following formulas: Fp=α×Iep+(1-α)×Ihp; Fc = α × Iec + (1-α) × Ihc; Fv=α×Iev+(1-α)×Ihv; Among them, F p 、F c and F v are the final impact severity of PM2.5, CO2 and VOCs, respectively; α is the weight coefficient of environmental impact on the final impact severity; (1-α) is the weight coefficient of human health impact on the final impact severity; Compare the severity of the final impact of pollutant gases and determine the priority of pollutant gas removal; A water spray tower is used to absorb and remove PM2.5 particles from the air, reducing the concentration of particulate matter in the air; a palladium catalyst is used to promote the oxidation reaction of VOCs, converting them into harmless carbon dioxide and water, reducing the impact of toxic gases on the environment; a sodium hydroxide solution is used to absorb CO2, and through a chemical reaction, CO2 is converted into sodium carbonate, reducing the concentration of carbon dioxide in the air; after the polluted gas is digested, the safety control mode unit adjusts the air volume and pressure according to the changes in the concentration of the polluted gas to ensure unobstructed air circulation and discharge the remaining polluted gas.

9. The intelligent control system for air volume and pressure according to claim 1, characterized in that: The data monitoring and analysis unit monitors data changes in real time and performs analysis; The model parameter dynamic adjustment unit adjusts the model parameters according to the feedback; The air volume and pressure control model unit optimizes air volume and pressure regulation based on the support vector regression algorithm, including: The data monitoring and analysis unit monitors changes in temperature, humidity, CO2 concentration, PM2.5 concentration, and VOCs concentration in the factory in real time, performs dynamic analysis, collects new sensor data, and calculates the current environmental conditions. The model parameter dynamic adjustment unit dynamically adjusts the parameters in the air volume and air pressure control model based on feedback from real-time monitoring data. The air volume and pressure control model unit is based on the support vector regression algorithm. Through training on historical data, it establishes a nonlinear relationship between air volume, air pressure and environmental parameters, and adjusts the air volume and air pressure in response to changes in temperature, humidity and pollutant gas concentration in the factory.

10. The intelligent control system for air volume and pressure according to claim 1, characterized in that: The parameter optimization unit optimizes the parameters in the control algorithm based on feedback; The model updating unit regularly updates the control model, including: The parameter optimization unit optimizes the parameters in the control algorithm based on real-time feedback data, including the PID control algorithm and the support vector regression algorithm, analyzes the impact of the control parameters on the air volume and pressure regulation, and adjusts the parameters through the genetic algorithm; The model update unit regularly updates the control model to ensure that the model can adapt to changes in the factory environment and new operational requirements, retrains the existing model using new monitoring data, and optimizes model performance by introducing new data features, adjusting the model structure, and updating training parameters.

Images