Self-cleaning type filtering unit based on synergistic effect of low-temperature plasma and catalyst

Abstract

The invention relates to the technical field of gas purification, in particular to a self-purification type filtering unit based on the synergistic effect of low-temperature plasma and a catalyst, which comprises a shell and further comprises a plasma generator, a filter element, a gas inlet pipe, a gas outlet pipe, a gas inlet pipe, a gas outlet pipe, a gas inlet pipe and a gas outlet pipe, the purification control unit is arranged on the outer wall of the shell and is used for controlling the air inlet flow of the self-purification type filtering unit; according to the self-purification type filtering unit based on the synergistic effect of the low-temperature plasma and the catalyst, the gas inlet flow is predicted according to the gas pollution coefficient, the external environment temperature and humidity and the electron density and the electron temperature of the plasma, and the gas inlet flow in the reaction chamber is collected in real time; and the real-time air inlet flow and the predicted air inlet flow are compared and analyzed to generate an adjustment and maintenance instruction, so that the purification effect of the self-purification type filtering unit based on the synergistic effect of the low-temperature plasma and the catalyst is ensured.

Description

Technical Field

[0001] The invention relates to the technical field of gas purification, in particular to a self-cleaning filter unit based on the synergistic effect of low-temperature plasma and a catalyst. Background Art

[0002] Low-temperature plasma technology is a highly efficient air purification technology that can decompose harmful substances in the air by generating active particles such as high-energy electrons, ions, and free radicals. The use of catalysts can further improve the degradation efficiency of pollutants. The low-temperature plasma filter unit is an advanced technology device for air purification and pollution control. It mainly eliminates pollutants in the air, such as volatile organic compounds (VOCs), nitrogen oxides, and particulate matter, through low-temperature plasma (LTP) technology.

[0003] The Chinese patent with publication number CN108325380A discloses a gas purification device with low-temperature plasma synergistic adsorption catalysis, which has good purification effect and no secondary pollution, including: flue shell, rectifying grid, dust filter component, low-temperature plasma reaction component, tail flue gas adsorption component and axial flow fan on the side of flue shell, the low-temperature plasma reaction component includes a plurality of plasma reaction tubes arranged in a certain manner, an arrangement fixing plate and a high-voltage AC power supply; the plasma reaction tube body is a cylindrical glass tube, the outer wall of which is tightly attached to a layer of mesh electrode, a metal inner electrode is placed at the axis, and two open baffles are arranged inside the cylindrical glass tube to allow flue gas to circulate, and a catalyst with a certain gap is filled between the two open baffles. Based on dielectric barrier discharge, low-temperature plasma is generated to react with VOCs, and the synergistic adsorption catalysis improves the removal efficiency and discharge stability.

[0004] As in the above application, the purification efficiency of existing low-temperature plasma-based purification and filtration equipment will be affected by the different degrees of gas pollution. If the air intake flow cannot be dynamically adjusted based on the degree of gas pollution and the working state of the plasma generator, it will easily cause incomplete gas purification or energy waste of the low-temperature plasma equipment, which is not conducive to energy conservation and emission reduction. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a self-cleaning filter unit based on the synergistic effect of low-temperature plasma and catalyst.

[0006] The present invention adopts the following technical scheme, a self-cleaning filter unit based on the synergistic effect of low-temperature plasma and catalyst, comprising a housing, a reaction chamber is arranged in the housing, and further comprising:

[0007] A plasma generator, wherein the plasma generator is disposed in the reaction chamber;

[0008] A purification control unit is arranged on the outer wall of the housing and is used to control the air intake flow of the self-cleaning filter unit. The purification control unit includes:

[0009] A data acquisition module collects historical purification training parameters of the self-cleaning filter unit. The historical purification training parameters are collected when the self-cleaning filter unit is in a state of reaching purification standards. The historical purification training parameters include comprehensive gas filtration parameters and intake air flow rate.

[0010] Model training module: Based on historical purification training parameters, train a machine learning model to predict the inlet air flow rate, collect real-time gas filtration comprehensive parameters, and predict the inlet air flow rate based on the trained machine learning model;

[0011] The analysis control module collects the real-time air intake flow in the reaction chamber, compares the real-time air intake flow with the predicted air intake flow, obtains the flow decay value, compares and analyzes the flow decay value with the preset flow decay gradient threshold, and generates adjustment and maintenance instructions.

[0012] As a further description of the above technical solution: the shell is provided with an air inlet and an air outlet at both ends along the length direction, and the reaction chamber is interconnected with the air inlet and the air outlet to form an airway for gas circulation.

[0013] As a further description of the above technical solution: a catalyst laying mechanism is also arranged in the reaction chamber, and the catalyst laying mechanism includes a rectangular laying frame rotatably arranged in the reaction chamber, a plurality of laying tubes are conductively connected in the laying frame, a plurality of blades distributed in a ring are arranged on the outer wall of the laying tube, and a catalyst outlet is conductively connected on the laying tube and located between two adjacent blades.

[0014] As a further description of the above technical solution: a first filter layer and a second filter layer are respectively arranged on both sides of the plasma generator in the reaction chamber.

[0015] As a further description of the above technical solution: the comprehensive gas filtration parameters include gas pollution coefficient, external environment temperature and humidity, and plasma electron density and electron temperature;

[0016] The parameters affecting the gas pollution coefficient include the concentration of volatile organic compounds in the gas, the concentration of PM2.5 in the gas, and the concentration of nitrogen oxides in the gas;

[0017] The expression of the gas pollution coefficient is:

[0018]

[0019] In the formula, QTwr is the gas pollution coefficient, YJnd is the concentration of volatile organic compounds in the gas, KLnd is the concentration of PM2.5 in the gas, DYnd is the concentration of nitrogen oxides in the gas, and is the weight factor, and Both are greater than 0.

[0020] As a further description of the above technical solution: the training method of the machine learning model for predicting the inlet and outlet air flow rate includes:

[0021] Convert the collected historical purification training parameters into a corresponding set of feature vectors;

[0022] Each group of feature vectors is used as the input of the machine learning model. The machine learning model takes the predicted intake flow corresponding to each group of gas filtration comprehensive parameters as the output, takes the intake flow actually corresponding to each group of gas filtration comprehensive parameters as the prediction target, and takes minimizing the loss function value of the machine learning model as the training target; training is stopped when the loss function value of the machine learning model is less than or equal to the preset target loss value.

[0023] The machine learning model is support vector machine regression, random forest regression or neural network regression.

[0024] As a further description of the above technical solution: the method for obtaining the flow decay value includes:

[0025] SBqt=SSqt-YCqt;

[0026] Where SBqt is the flow decay value, SSqt is the real-time intake flow, and YCqt is the predicted intake flow.

[0027] As a further description of the above technical solution: the adjustment and maintenance instruction includes a first flow adjustment instruction, a second flow adjustment instruction, a first alarm maintenance instruction and a second alarm maintenance instruction, wherein the first flow adjustment instruction is to increase the intake flow, and the second flow adjustment instruction is to reduce the intake flow.

[0028] As a further description of the above technical solution: the method of comparing and analyzing the flow decay value with the preset flow decay gradient threshold to generate an adjustment maintenance instruction includes:

[0029] Preset flow decay gradient threshold LL1<LL2<0<LL3<LL4;

[0030] When LL2<SBqt<LL3, no adjustment maintenance instruction is generated;

[0031] When LL1≤SBqt≤LL2, a first flow adjustment instruction is generated;

[0032] When SBqt<LL1, the first alarm maintenance instruction is generated;

[0033] When LL3≤SBqt≤LL4, a second flow rate adjustment instruction is generated;

[0034] When SBqt>LL4, a second warning maintenance instruction is generated.

[0035] As a further description of the above technical solution: the method for obtaining the electron density of the plasma includes:

[0036] The optical probe is mounted on the observation window of the reaction chamber so that the probe can directly collect the light emitted by the plasma;

[0037] Connecting the optical probe to the fiber optic spectrometer, starting the spectrometer to collect spectrum data of the plasma in real time, and selecting a predetermined characteristic spectrum line;

[0038] Real-time recording of the full width at half maximum of characteristic spectral lines;

[0039] The electron density is calculated by the formula:

[0040] Δλ 1 / 2 =2ω·n e 2 / 3 ;

[0041] In the formula, Δλ 1 / 2 is the full width at half maximum of the characteristic spectral line, ω is the Stark broadening coefficient, n e is the electron density in the plasma.

[0042] Beneficial effects:

[0043] The self-cleaning filter unit based on the synergistic effect of low-temperature plasma and catalyst provided by the present invention predicts the intake flow rate according to the gas pollution coefficient, the temperature and humidity of the external environment, the electron density and the electron temperature of the plasma, and collects the real-time intake flow rate in the reaction chamber, compares the real-time intake flow rate with the predicted intake flow rate, obtains the flow decay value, compares and analyzes the flow decay value with the preset flow decay gradient threshold, and generates an adjustment and maintenance instruction, thereby ensuring the purification effect of the self-cleaning filter unit based on the synergistic effect of low-temperature plasma and catalyst, realizing the optimization of the intake flow rate of the filter unit by using a machine learning algorithm, and achieving the optimal balance between energy consumption and purification efficiency;

[0044] Furthermore, when the self-cleaning filtration unit based on the synergistic effect of low-temperature plasma and catalyst is used, first, a suitable corresponding catalyst is selected according to the type of pollution-free gas and the purification requirements, and the catalyst is added to the layout frame. Then, the catalyst is discharged through the catalyst outlet opened on the layout pipe and enters the reaction chamber to contact and react with the polluted gas. In use, when the polluted gas enters the reaction chamber, it will contact the blade set on the layout pipe. The gas blows the blade to rotate, thereby driving the entire layout frame to rotate, so that the catalyst sprayed through the catalyst outlet opened on the layout pipe is more evenly dispersed in the reaction chamber, thereby improving its catalytic effect. In addition, through the setting of the blade, the polluted gas discharged into the reaction chamber can be slowed down, thereby increasing its contact time with the catalyst, and further improving its pollutant degradation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The present invention will be further explained below in conjunction with the accompanying drawings and embodiments:

[0046] Figure 1 A schematic cross-sectional view of a self-cleaning filter unit based on the synergistic effect of low-temperature plasma and catalyst provided by the present invention;

[0047] Figure 2 The present invention provides Figure 1 A magnified view of area A in;

[0048] Figure 3 A schematic diagram of a module of a purification control unit provided by the present invention;

[0049] Figure 4 The present invention provides a flow chart of a method for obtaining the electron density of plasma.

[0050] Figure numerals: 1, outer shell; 11, air inlet; 12, air outlet; 13, reaction chamber; 131, plasma generator; 132, first filter layer; 133, second filter layer; 1341, laying frame; 1342, laying pipe; 1343, blade; 1344, catalyst outlet. DETAILED DESCRIPTION

[0051] In order to make the technical means, creative features, objectives and effects of the present invention easy to understand, the present invention is further described below with reference to specific diagrams. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0052] Example 1

[0053] See also Figure 1-Figure 3The embodiment of the present invention provides a technical solution: a self-cleaning filter unit based on the synergistic effect of low-temperature plasma and catalyst, comprising a housing 1, a reaction chamber 13 is arranged in the housing 1, an air inlet 11 and an air outlet 12 are respectively arranged at both ends of the housing 1 along the length direction, and the reaction chamber 13 is interconnected with the air inlet 11 and the air outlet 12 to form an airway for gas circulation. When in use, polluted gas enters through the air inlet 11, enters the reaction chamber 13 for purification, and is discharged through the air outlet 12 to complete air purification, and also includes:

[0054] The plasma generator 131 is disposed in the reaction chamber 13 . A first filter layer 132 and a second filter layer 133 are disposed on both sides of the plasma generator 131 in the reaction chamber 13 .

[0055] A purification control unit is arranged on the outer wall of the housing 1 (the purification control unit is not shown in the figure) and is used to control the air intake flow of the self-cleaning filter unit. The purification control unit includes:

[0056] A data acquisition module collects historical purification training parameters of the self-cleaning filter unit. The historical purification training parameters are collected when the self-cleaning filter unit is in a state of reaching purification standards. The historical purification training parameters include comprehensive gas filtration parameters and intake air flow rate.

[0057] The comprehensive gas filtration parameters include gas pollution coefficient, external environment temperature and humidity, plasma electron density and electron temperature;

[0058] Specifically, the gas pollution coefficient is a basis for judging the degree of gas pollution. The smaller the gas pollution coefficient is, the lower the degree of gas pollution is. Correspondingly, when passing through the self-cleaning filter unit, the faster the purification efficiency is, and the intake flow rate can be controlled to increase, and vice versa.

[0059] The parameters affecting the gas pollution coefficient include the concentration of volatile organic compounds in the gas, the concentration of PM2.5 in the gas, and the concentration of nitrogen oxides in the gas;

[0060] Specifically, obtaining the concentration of volatile organic compounds (VOCs) in the gas is an important step in evaluating the degree of gas pollution and the purification effect of the self-cleaning filter unit. Commonly used methods include gas chromatography, photoionization detection, mass spectrometry and electrochemical sensors;

[0061] The concentration of PM2.5 in the gas can be obtained by using the laser particle counting method, which uses the principle of laser scattering to calculate the concentration of particles by measuring the scattering intensity of laser by particles in the gas.

[0062] The method for obtaining the concentration of nitrogen oxides in the gas is chemiluminescence, which uses the reaction of NO and ozone to generate excited nitrogen dioxide. NO2 emits light when returning to the ground state. The intensity of the light is proportional to the concentration of NO. By pre-establishing the corresponding relationship between the intensity of light and the concentration of NO, the concentration of nitrogen oxides in the gas can be obtained.

[0063] The expression of the gas pollution coefficient is:

[0064]

[0065] In the formula, QTwr is the gas pollution coefficient, YJnd is the concentration of volatile organic compounds in the gas, KLnd is the concentration of PM2.5 in the gas, DYnd is the concentration of nitrogen oxides in the gas, and is the weight factor, and Both are greater than 0.

[0066] It should be noted that the weight factor is set to quantify each data to obtain a specific value for subsequent comparison. The size of the weight factor depends on the number of gas filtration comprehensive parameters and the technical personnel in this field preliminarily set the corresponding weight coefficient for each set of gas filtration comprehensive parameters.

[0067] Specifically, the higher the concentration of volatile organic compounds in the gas, the greater the degree of gas pollution, and vice versa. The higher the concentration of nitrogen oxides in the gas, the greater the degree of gas pollution, and vice versa. The higher the concentration of PM2.5 in the gas, the greater the degree of gas pollution, and vice versa.

[0068] Model training module: Based on historical purification training parameters, train a machine learning model to predict the inlet air flow rate, collect real-time gas filtration comprehensive parameters, and predict the inlet air flow rate based on the trained machine learning model;

[0069] The training method of the machine learning model for predicting the inlet and outlet air flow rate includes:

[0070] Convert the collected historical purification training parameters into a corresponding set of feature vectors;

[0071] Each group of feature vectors is used as the input of the machine learning model. The machine learning model takes the predicted intake flow corresponding to each group of gas filtration comprehensive parameters as the output, takes the intake flow actually corresponding to each group of gas filtration comprehensive parameters as the prediction target, and takes minimizing the loss function value of the machine learning model as the training target; training is stopped when the loss function value of the machine learning model is less than or equal to the preset target loss value.

[0072] The machine learning model may be one of support vector machine regression, random forest regression or neural network regression models.

[0073] The loss function value of the machine learning model is the mean square error.

[0074] Mean square error is one of the commonly used loss functions. The model is trained with minimization as the goal, so that the machine learning model can better fit the data, thereby improving the performance and accuracy of the model.

[0075] In the loss function, MSE is the loss function value of the machine learning model, x is the feature vector group number; m is the number of feature vector groups; y is the number of feature vector groups; x is the intake flow rate corresponding to the xth group of eigenvectors, is the intake flow corresponding to the x-th group of characteristic vectors in real time.

[0076] Other model parameters of the machine learning model, target loss value, optimization algorithm, training set test set validation set ratio, and loss function optimization are all achieved through actual engineering implementation and continuous experimental tuning.

[0077] The analysis control module collects the real-time air intake flow in the reaction chamber 13, compares the real-time air intake flow with the predicted air intake flow, obtains the flow decay value, compares and analyzes the flow decay value with the preset flow decay gradient threshold, and generates an adjustment maintenance instruction.

[0078] The method for obtaining the flow decay value includes:

[0079] SBqt=SSqt-YCqt;

[0080] Where SBqt is the flow decay value, SSqt is the real-time intake flow, and YCqt is the predicted intake flow.

[0081] The adjustment and maintenance instructions include a first flow adjustment instruction, a second flow adjustment instruction, a first alarm maintenance instruction and a second alarm maintenance instruction, wherein the first flow adjustment instruction is to increase the intake flow, and the second flow adjustment instruction is to reduce the intake flow. It should be noted that the intake flow is adjusted by changing the size of the valve opening in the air intake port 11.

[0082] The method of comparing and analyzing the flow decay value with a preset flow decay gradient threshold and generating an adjustment maintenance instruction includes:

[0083] Preset flow decay gradient threshold LL1<LL2<0<LL3<LL4;

[0084] When LL2<SBqt<LL3, it means that the flow decay value is within the allowable error range, so the sewage adjusts the intake flow and does not generate an adjustment maintenance instruction;

[0085] When LL1≤SBqt≤LL2, it means that the real-time intake air flow rate is less than the predicted intake air flow rate, which means that the real-time intake air flow rate is small, so the real-time intake air flow rate needs to be increased, and thus a first flow rate adjustment instruction is generated;

[0086] When SBqt<LL1, a first alarm maintenance instruction is generated, indicating that the real-time intake flow rate is far less than the predicted intake flow rate, and flow adjustment cannot be achieved simply by adjusting the real-time intake flow rate, and maintenance is required;

[0087] When LL3≤SBqt≤LL4, it means that the real-time intake air flow rate is greater than the predicted intake air flow rate, which means that the real-time intake air flow rate is large, so the real-time intake air flow rate needs to be reduced, and thus a second flow rate adjustment instruction is generated;

[0088] When SBqt>LL4, a second alarm maintenance instruction is generated, indicating that the real-time intake flow rate is much larger than the predicted intake flow rate, and flow adjustment cannot be achieved simply by adjusting the real-time intake flow rate, and maintenance is required.

[0089] In this embodiment, the intake air flow rate is predicted based on the gas pollution coefficient, the external environment temperature and humidity, the electron density and electron temperature of the plasma, and by collecting the real-time intake air flow rate in the reaction chamber 13, the real-time intake air flow rate is compared with the predicted intake air flow rate, and the flow decay value is obtained. The flow decay value is compared and analyzed with the preset flow decay gradient threshold, and an adjustment and maintenance instruction is generated, thereby ensuring the purification effect of the self-cleaning filter unit based on the synergistic effect of low-temperature plasma and catalyst, realizing the optimization of the intake air flow rate of the self-cleaning filter unit by using a machine learning algorithm, and achieving the optimal balance between energy consumption and purification efficiency.

[0090] Example 2

[0091] Reference Figure 1-Figure 2 In this embodiment, a catalyst placement mechanism is added on the basis of the above embodiment. A catalyst placement mechanism is also provided in the reaction chamber 13, and the catalyst placement mechanism includes a rectangular placement frame 1341 rotatably provided in the reaction chamber 13, a plurality of placement tubes 1342 are conductively connected in the placement frame 1341, a plurality of annularly distributed blades 1343 are provided on the outer wall of the placement tube 1342, and a catalyst outlet 1344 is conductively connected on the placement tube 1342 and located between two adjacent blades 1343.

[0092] Specifically, by setting up the catalyst arrangement mechanism, when the gas is purified by the self-cleaning filter unit, the catalyst can be added through the catalyst arrangement mechanism. The use of the catalyst can further improve the degradation efficiency of the pollutants.

[0093] In this embodiment, when in use, firstly, according to the type of pollution-free gas and the purification requirements, a suitable corresponding catalyst is selected, and the catalyst is added to the layout frame 1341. It should be noted that the layout frame 1341 is a rectangular hollow structure. Then the catalyst is discharged through the catalyst outlet 1344 opened on the layout pipe 1342 and enters the reaction chamber 13 to contact and react with the polluted gas. In use, when the polluted gas enters the reaction chamber 13, it will contact the blade 1343 set on the layout pipe 1342. The blade 1343 is rotated by the gas blowing, thereby driving the entire layout frame 1341 to rotate, so that the catalyst sprayed through the catalyst outlet 1344 opened on the layout pipe 1342 is more evenly dispersed in the reaction chamber 13, thereby improving its catalytic effect. In addition, through the setting of the blade 1343, the polluted gas discharged into the reaction chamber 13 can be slowed down to increase its contact time with the catalyst, thereby further improving its pollutant degradation efficiency.

[0094] Example 3

[0095] Reference Figure 4 Based on the above embodiments, this embodiment further adds a method for obtaining the electron density of the plasma.

[0096] The method for obtaining the electron density of the plasma includes:

[0097] An optical probe is mounted on the observation window of the reaction chamber 13 so that the probe can directly collect the light emitted by the plasma;

[0098] Connect the optical probe to the fiber optic spectrometer, start the spectrometer to collect the spectral data of the plasma in real time, select the predetermined characteristic spectral line, for example, select the Line, wavelength 656.3nm;

[0099] Real-time recording of the full width at half maximum of characteristic spectral lines;

[0100] The electron density is calculated by the formula:

[0101] Δλ 1 / 2 =2ω·n e 2 / 3 ;

[0102] In the formula, Δλ 1 / 2 is the full width at half maximum of the characteristic spectral line, ω is the Stark broadening coefficient, n e is the electron density in the plasma.

[0103] It should be noted that the method for obtaining the electron temperature of the plasma is the optical emission spectroscopy (OES) method, and the optical emission spectroscopy method calculates the electron temperature by analyzing the relative intensities of different characteristic spectral lines.

[0104] In this embodiment, the working efficiency of the plasma generator 131 in the reaction chamber 13 is reflected by obtaining the electron density and electron temperature of the plasma in the reaction chamber 13, and the intake flow rate is optimized based on the working efficiency of the plasma generator 131, thereby realizing dynamic adjustment of the intake flow rate and further ensuring the gas purification effect.

[0105] The basic principles, main features and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and the description in the specification are only to illustrate the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A self-cleaning filter unit based on the synergistic effect of low-temperature plasma and catalyst, comprising a housing (1), wherein a reaction chamber (13) is arranged in the housing (1), characterized in that: Also includes: A plasma generator (131), wherein the plasma generator (131) is arranged in the reaction chamber (13); A purification control unit is used to control the air intake flow of the self-cleaning filter unit, and the purification control unit includes: A data acquisition module collects historical purification training parameters of the self-cleaning filter unit. The historical purification training parameters are collected when the self-cleaning filter unit is in a state of reaching purification standards. The historical purification training parameters include comprehensive gas filtration parameters and intake air flow rate. Model training module: Based on historical purification training parameters, train a machine learning model to predict the inlet air flow rate, collect real-time gas filtration comprehensive parameters, and predict the inlet air flow rate based on the trained machine learning model; The analysis control module collects the real-time air intake flow rate in the reaction chamber (13), compares the real-time air intake flow rate with the predicted air intake flow rate, obtains the flow decay value, compares and analyzes the flow decay value with a preset flow decay gradient threshold, and generates an adjustment maintenance instruction.

2. The self-cleaning filtration unit based on the synergistic effect of low-temperature plasma and catalyst according to claim 1, characterized in that: The housing (1) is provided with an air inlet (11) and an air outlet (12) at both ends along the length direction, and the reaction chamber (13) is interconnected with the air inlet (11) and the air outlet (12) to form an air passage for gas circulation.

3. The self-cleaning filtration unit based on the synergistic effect of low-temperature plasma and catalyst according to claim 2, characterized in that: A catalyst placement mechanism is also provided in the reaction chamber (13), and the catalyst placement mechanism comprises a rectangular placement frame (1341) rotatably arranged in the reaction chamber (13), a plurality of placement tubes (1342) are conductively connected in the placement frame (1341), a plurality of blades (1343) distributed in an annular shape are provided on the outer wall of the placement tube (1342), and a catalyst outlet (1344) is conductively connected on the placement tube (1342) and located between two adjacent blades (1343).

4. The self-cleaning filtration unit based on the synergistic effect of low-temperature plasma and catalyst according to claim 2, characterized in that: A first filter layer (132) and a second filter layer (133) are respectively arranged on both sides of the plasma generator (131) in the reaction chamber (13).

5. The self-cleaning filtration unit based on the synergistic effect of low-temperature plasma and catalyst according to claim 4, characterized in that: The comprehensive gas filtration parameters include gas pollution coefficient, external environment temperature and humidity, and plasma electron density and electron temperature; The parameters affecting the gas pollution coefficient include the concentration of volatile organic compounds in the gas, the concentration of PM2.5 in the gas, and the concentration of nitrogen oxides in the gas; The expression of the gas pollution coefficient is: In the formula, QTwr is the gas pollution coefficient, YJnd is the concentration of volatile organic compounds in the gas, KLnd is the concentration of PM2.5 in the gas, DYnd is the concentration of nitrogen oxides in the gas, and is the weight factor, and Both are greater than 0.

6. The self-cleaning filtration unit based on the synergistic effect of low-temperature plasma and catalyst according to claim 1, characterized in that: The training method of the machine learning model for predicting the inlet and outlet air flow rate includes: Convert the collected historical purification training parameters into a corresponding set of feature vectors; Each set of feature vectors is used as the input of the machine learning model. The machine learning model uses the predicted intake flow corresponding to each set of gas filtration comprehensive parameters as the output, uses the intake flow actually corresponding to each set of gas filtration comprehensive parameters as the prediction target, and minimizes the loss function value of the machine learning model as the training target; when the loss function value of the machine learning model is less than or equal to the preset target loss value, the training is stopped; The machine learning model is support vector machine regression, random forest regression or neural network regression.

7. The self-cleaning filtration unit based on the synergistic effect of low-temperature plasma and catalyst according to claim 1, characterized in that: The method for obtaining the flow decay value includes: SBqt=SSqt-YCqt; Where SBqt is the flow decay value, SSqt is the real-time intake flow, and YCqt is the predicted intake flow.

8. The self-cleaning filtration unit based on the synergistic effect of low-temperature plasma and catalyst according to claim 7, characterized in that: The adjustment and maintenance instruction includes a first flow adjustment instruction, a second flow adjustment instruction, a first alarm maintenance instruction and a second alarm maintenance instruction, wherein the first flow adjustment instruction is to increase the intake flow, and the second flow adjustment instruction is to reduce the intake flow.

9. The self-cleaning filtration unit based on the synergistic effect of low-temperature plasma and catalyst according to claim 8, characterized in that: The method of comparing and analyzing the flow decay value with a preset flow decay gradient threshold to generate an adjustment maintenance instruction includes: Preset flow decay gradient threshold LL1<LL2<0<LL3<LL4; When LL2<SBqt<LL3, no adjustment maintenance instruction is generated; When LL1≤SBqt≤LL2, a first flow adjustment instruction is generated; When SBqt<LL1, the first alarm maintenance instruction is generated; When LL3≤SBqt≤LL4, a second flow rate adjustment instruction is generated; When SBqt>LL4, a second warning maintenance instruction is generated.

10. The self-cleaning filtration unit based on the synergistic effect of low-temperature plasma and catalyst according to claim 5, characterized in that: The method for obtaining the electron density of the plasma includes: Installing an optical probe on an observation window of the reaction chamber (13) so that the probe collects light emitted by the plasma; Connecting the optical probe to the fiber optic spectrometer, starting the spectrometer to collect spectrum data of the plasma in real time, and selecting a predetermined characteristic spectrum line; Real-time recording of the full width at half maximum of characteristic spectral lines; The electron density is calculated by the formula: Dl 1 / 2 =2ω·n e 2 / 3 ; In the formula, Δλ 1 / 2 is the full width at half maximum of the characteristic spectral line, ω is the Stark broadening coefficient, n e is the electron density in the plasma.

Images