A control method for bipolar double-reaction-zone pulse air purification

Abstract

The present application relates to air purification technical field, specifically to a kind of bipolar double reaction zone pulse air purification control method, comprising the following steps: step one, hardware layout;Step two, model loading;Step three, information collection;Step four, parameter adjustment;Step five, monitoring alarm.The present application is based on simulation database by machine learning model, real-time analyzes the relevance of ozone concentration and each pulse parameter and dynamically adjusts pulse parameter, realizes the real-time dynamic adjustment of ozone concentration, reduces the risk of excessive ozone due to uncontrolled pulse discharge, and further enables the ozone generated to be more comprehensive decomposition, while also achieving dynamic optimal balance of purification efficiency and safety.

Description

Technical Field

[0001] This invention relates to the field of air purification technology, specifically to a control method for bipolar dual-reaction zone pulse air purification. Background Technology

[0002] The bipolar dual-reaction-zone pulse air purification control method is a technical solution that achieves high-efficiency air purification through the synergistic effect of a bipolar pulse high-voltage power supply and a dual-reaction-zone reactor. It treats gaseous pollutants by applying positive and negative pulses to the positive and negative pulse reaction zones respectively, achieving differentiated removal effects for different pollutant components. Simultaneously, by synchronously regulating the positive and negative pulse parameters through the control system, it improves purification efficiency while reducing the generation of secondary pollutants, lowering the environmental impact of noise and vibration, and reducing maintenance costs by simplifying the control system. Ultimately, it achieves highly efficient, low-pollution, and easily scalable gaseous pollutant treatment.

[0003] The existing bipolar dual-reaction-zone pulse air purification control method drives the dual-reaction-zone reactor with a bipolar pulse high-voltage power supply, applies a positive pulse to the positive pulse reaction zone and a negative pulse to the negative pulse reaction zone to achieve synergistic treatment of gaseous pollutants. At the same time, it also adjusts the ozone concentration through a post-treatment device to control ozone generation and reduce secondary pollution.

[0004] While the above methods can treat gaseous pollutants to a certain extent, they do not have the ability to dynamically adjust ozone concentration in real time because the adjustment of ozone concentration is done through a post-treatment device. This may result in ozone not being completely decomposed.

[0005] In summary, the problem that ozone concentration may not be completely decomposed by the existing technology of adjusting ozone concentration through post-treatment device has become an urgent problem to be solved in this field. Therefore, it is necessary to propose a control method for bipolar dual-reaction zone pulse air purification. Summary of the Invention

[0006] To address the aforementioned issues, this invention provides a control method for bipolar dual-reaction zone pulse air purification. Based on a simulation database and using a machine learning model, the method analyzes the correlation between ozone concentration and various pulse parameters in real time and dynamically adjusts the pulse parameters. This achieves real-time dynamic regulation of ozone concentration, reduces the risk of excessive ozone generation due to uncontrollable pulse discharge, and enables more comprehensive decomposition of generated ozone. Simultaneously, it achieves a dynamic optimal balance between purification efficiency and safety.

[0007] To achieve the above objectives, the technical solution of the present invention is as follows: A control method for bipolar dual-reaction zone pulse air purification, comprising the following steps: Step 1, Hardware Setup: Multi-parameter sensor arrays are installed at both the inlet and outlet of the reactor. A dust collection zone and a pre-ionization zone are set up in the reactor. The pre-ionization zone consists of needle electrodes and mesh electrodes, and the dust collection zone consists of several parallel plate electrodes. The pre-ionization zone is electrically connected to the pulse power supply, and the dust collection zone is electrically connected to the DC power supply.

[0008] Step 2, Model Loading: The machine learning model is trained using gradient boosting decision trees and then loaded.

[0009] Step 3, Information Collection: A multi-parameter sensor array is used to collect information on PM2.5 concentration, PM10 concentration, TVOC concentration, temperature and humidity, and ozone concentration. All collected information is integrated to form environmental data. The environmental data is then filtered and denoised to extract its feature values. These feature values ​​are imported into a machine learning model for analysis. The model determines the specific type of pollution source and predicts the trend of pollutant concentration changes and the pollution load in the dust collection area within the next 1-3 minutes. Based on these predictions, a simulation database is constructed.

[0010] Step 4, parameter adjustment: Based on the prediction results of the machine learning model, the purification efficiency and energy consumption of the reactor are weighted and allocated. The drive signal of the pulse generation circuit in the reactor is modified according to the weight allocation result, and a safe threshold for ozone concentration is set to control the ozone concentration below the set safe threshold.

[0011] Step 5, Monitoring and Alarm: Monitor the pollution load and purification efficiency of the dust collection area in real time, set the pollution load threshold for the dust collection area, and send an alarm message to the staff when the pollution load of the dust collection area exceeds the pollution load threshold, or when the purification efficiency of the dust collection area decreases by more than 5% within 3-5 minutes under the same pulse parameters.

[0012] Furthermore, in step one, a solid-state circuit breaker is connected in series at the output terminal of the pulse power supply, and a varistor is connected in parallel.

[0013] Furthermore, in step one, the reactor also integrates IGBT modules, which in turn integrate current and temperature sensors. The current sensor monitors the collector current of the IGBT module in real time to obtain its current data, and the temperature sensor monitors the casing temperature of the IGBT module in real time to obtain its temperature data. The current and temperature data of the IGBT module are input into a machine learning model for analysis to obtain integrated data. Based on this integrated data, an IGBT aging prediction model is established. After the operator sets the lifespan threshold for the IGBT module, the IGBT aging prediction model predicts the remaining lifespan of the IGBT module. When the predicted result is lower than the lifespan threshold, an alarm message is sent to the operator.

[0014] Furthermore, in step one, the multi-parameter sensor array includes a PM2.5 concentration sensor, a PM10 concentration sensor, a TVOC concentration sensor, a temperature and humidity sensor, and an ozone concentration sensor, and the sampling frequency of the multi-parameter sensor array is 1-2 seconds / time.

[0015] Furthermore, in step three, the machine learning model updates the prediction results by recording them into the simulation database each time it produces a prediction.

[0016] Furthermore, in step four, when modifying the drive signal of the pulse generation circuit, the waveform used is one of square wave, trapezoidal wave, and resonant waveform.

[0017] Furthermore, in step four, the machine learning model recommends pulse parameters to staff based on the simulation database, and adjusts the pulse parameters according to the real-time monitored ozone concentration. When the ozone concentration approaches the safety threshold, the input pulse voltage and frequency are reduced.

[0018] Furthermore, in step four, the pulse parameters include the DC bias voltage V1 of the dust collection zone, and the pulse voltage V2, frequency F2, and pulse width W2 of the pre-ionization zone.

[0019] Furthermore, in step four, when modifying the drive signal of the pulse generation circuit, the modification logic is as follows: When particulate matter is the main pollutant, the pulse parameters of the pre-ionization zone are optimized to improve the charging efficiency of particulate matter.

[0020] When VOCs are the main pollutant type, the pulse parameters of the pre-ionization zone are optimized to enhance the generation of free radicals and the catalytic decomposition of VOCs by non-equilibrium plasma.

[0021] Furthermore, in steps one and five, when sending alarm information to staff, an audible and visual alarm is used, and an alarm SMS is sent to the staff's terminal at the same time.

[0022] The above approach has the following beneficial effects: 1. This invention uses a machine learning model based on a simulation database to analyze the correlation between ozone concentration and various pulse parameters in real time and dynamically adjust the pulse parameters. This achieves real-time dynamic adjustment of ozone concentration, reduces the risk of excessive ozone generation due to uncontrollable pulse discharge, and enables the generated ozone to be decomposed more comprehensively. At the same time, it also achieves a dynamic optimal balance between purification efficiency and safety.

[0023] 2. This invention identifies the specific type of pollution source through a multi-parameter sensor array and machine learning model, and predicts the pollutant change trend and pollution load of the dust collection area in the next 1-3 minutes, so that the purification strategy can be adjusted in advance, thereby reducing the occurrence of purification delays caused by sudden pollution.

[0024] 3. This invention, by predicting and providing early warning of aging of IGBT modules, enables staff to understand the specific condition of the IGBT modules in a timely manner, allowing for timely repair or replacement, thereby reducing unplanned downtime and lowering maintenance costs.

[0025] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the method steps of the bipolar dual-reaction zone pulse air purification control method of the present invention.

[0027] Figure 2 This is a flowchart illustrating the information acquisition step in the control method for bipolar dual-reaction zone pulse air purification of the present invention.

[0028] Figure 3 This is a flowchart illustrating the monitoring and alarm steps in the control method for bipolar dual-reaction zone pulse air purification of the present invention.

[0029] Figure 4 This is a schematic diagram illustrating the steps of establishing an IGBT aging prediction model in the bipolar dual-reaction zone pulse air purification control method of the present invention. Detailed Implementation

[0030] The following detailed description illustrates the specific implementation method: Implementation, for example, attached Figure 1 As shown: A control method for bipolar dual-reaction zone pulse air purification includes the following steps: Step 1, Hardware Deployment: First, deploy multi-parameter sensor arrays at the inlet and outlet of the reactor. Within the reactor, establish a dust collection zone consisting of several parallel-arranged flat plate electrodes (8-12 plates, 10 in this embodiment, electrode spacing 5-10cm, 8cm in this embodiment) and a pre-ionization zone consisting of needle-shaped and mesh electrodes (needle electrode diameter 0.5-1mm, 0.8mm in this embodiment, mesh electrode aperture 1-3mm, 2mm in this embodiment). Connect the pre-ionization zone to a pulse power supply, and the dust collection zone to a DC power supply. Connect a solid-state circuit breaker in series and a varistor in parallel at the output of the pulse power supply (varistor nominal voltage 500-1000V, 800V in this embodiment). The multi-parameter sensor array includes PM2.5 concentration sensors, PM10 concentration sensors, TVOC concentration sensors, temperature and humidity sensors, and ozone concentration sensors, with a sampling frequency of 1 second / sample.

[0031] Step 2, Model Loading: While deploying the hardware, staff simultaneously load the machine learning model, which is trained using a gradient boosting decision tree.

[0032] like Figure 4 As shown, in this process, workers integrate IGBT modules into the reactor, along with current and temperature sensors. The current sensor monitors the collector current of the IGBT module in real time, obtaining its current data, while the temperature sensor monitors the casing temperature of the IGBT module in real time, obtaining its temperature data. The current and temperature data are then input into a machine learning model for analysis, resulting in integrated data. Based on this integrated data, an IGBT aging prediction model is established. After setting the lifespan threshold for the IGBT module, the IGBT aging prediction model predicts the remaining lifespan of the IGBT module. When the predicted result is lower than the lifespan threshold, an alarm is sent to the workers via audible and visual alarms, and an alarm SMS is simultaneously sent to the workers' terminals.

[0033] Step 3, Information Collection: such as Figure 2As shown, after the model is loaded, a multi-parameter sensor array deployed at the reactor inlet and outlet begins to collect information on PM2.5 concentration, PM10 concentration, TVOC concentration, temperature and humidity, and ozone concentration. All collected information is integrated to form environmental data. This data is then filtered and denoised, and feature values ​​are extracted. These feature values ​​are then imported into a machine learning model for analysis. The model determines the specific type of pollution source (e.g., particulate matter or VoCs) and predicts the pollutant concentration trends and pollution load in the dust collection area within the next 1-3 minutes. Based on these predictions, a simulation database is built. Each time the machine learning model generates a prediction, it updates the simulation database by recording the result.

[0034] Step four, parameter adjustment: After predicting the trend of pollutant concentration changes, the staff, based on the prediction results of the machine learning model, weighted the reactor's purification efficiency and energy consumption (e.g., purification efficiency weighted at 0.7, energy consumption weighted at 0.3). After completing the weighting, the drive signal of the pulse generation circuit in the reactor was modified according to the weighting results, and a safe threshold for ozone concentration was set to ensure that the ozone concentration remained below the set safe threshold. When modifying the drive signal of the pulse generation circuit, the waveform used is one of square wave, trapezoidal wave, or resonant waveform; in this embodiment, a trapezoidal wave is selected.

[0035] During this process, the machine learning model also recommends pulse parameters to staff based on a simulation database. These pulse parameters are adjusted according to the real-time monitored ozone concentration; for example, when the ozone concentration approaches a safe threshold, both the input pulse voltage and frequency are reduced. The pulse parameters include the DC bias voltage V1 in the dust collection zone, and the pulse voltage V2, frequency F2, and pulse width W2 in the pre-ionization zone.

[0036] Specifically, the logic for modifying the drive signal of the pulse generation circuit is as follows: When the pollutant type is mainly particulate matter, the charging efficiency of particulate matter can be improved by optimizing the pulse parameters of the pre-ionization zone (by increasing the pulse voltage V2 and extending the pulse width W2, while keeping the DC bias voltage V1 of the dust collection zone stable, so as to enhance the electric field strength and ion density and improve the charging efficiency).

[0037] When the pollutant type is mainly VOCs, the generation of free radicals and the catalytic decomposition of VOCs by non-equilibrium plasma can be enhanced by optimizing the pulse parameters of the pre-ionization zone (by increasing the pulse frequency F2 of the pre-ionization zone and switching the pulse waveform to a resonant waveform to improve the catalytic decomposition efficiency).

[0038] Step 5, monitoring and alarm: such as Figure 3 As shown, after the staff sets the pollution load threshold of the dust collection area, the machine learning model monitors the pollution load and purification efficiency of the dust collection area in real time. When the pollution load of the dust collection area exceeds the pollution load threshold, or when the purification efficiency of the dust collection area decreases by more than 5% within 3 minutes under the same pulse parameters, the system will send an alarm message to the staff through an audible and visual alarm, and at the same time send an alarm SMS to the staff's terminal.

[0039] This method uses a machine learning model based on a simulation database to analyze the correlation between ozone concentration and various pulse parameters in real time and dynamically adjust the pulse parameters. This achieves real-time dynamic regulation of ozone concentration, reduces the risk of excessive ozone generation due to uncontrollable pulse discharge, and enables the generated ozone to be decomposed more comprehensively. At the same time, it also achieves a dynamic optimal balance between purification efficiency and safety.

[0040] In this embodiment, the method of this embodiment is used as the experimental group, and the existing bipolar dual-reaction zone pulse air purification control method is used as the control group for comparative experiment.

[0041] The initial background pollutant concentrations were controlled at PM2.5 = 500 μg / m³, TVOC = 2 mg / m³, and the ambient temperature was 25 ± 1℃. The following experimental table was obtained: Table 1 Experimental Tables As shown in the table, the peak ozone concentration, duration of ozone exceedance, and average ozone concentration in the experimental group were all lower than those in the control group, indicating that the ozone concentration was successfully stabilized. Furthermore, the PM2.5 purification time, unit purification energy consumption, and peak IGBT temperature in the experimental group were all lower than those in the control group. Meanwhile, the TVOC purification efficiency and clean air output ratio in the experimental group were higher than those in the control group. It can be concluded that the method is superior to existing technologies in terms of ozone concentration control and air purification effect.

[0042] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A control method for bipolar dual-reaction zone pulse air purification, characterized in that, Includes the following steps: Step 1, Hardware Deployment: Multi-parameter sensor arrays are deployed at both the inlet and outlet of the reactor, and a dust collection zone and a pre-ionization zone are established in the reactor. The pre-ionization zone consists of needle-shaped electrodes and mesh electrodes; the dust collection zone consists of several parallel-arranged plate electrodes, and the pre-ionization zone is electrically connected to the pulse power supply, while the dust collection zone is electrically connected to the DC power supply. Step 2, Model Loading: The machine learning model is trained using gradient boosting decision trees and then loaded. Step 3, Information Collection: The PM2.5 concentration, PM10 concentration, TVOC concentration, temperature and humidity, and ozone concentration of the gas are collected through a multi-parameter sensor array. All the collected information is integrated to form environmental data. The environmental data is then filtered and denoised to extract its feature values. The feature values ​​are imported into the machine learning model, which analyzes them to determine the specific type of pollution source. The model also predicts the trend of pollutant concentration changes and the pollution load of the dust collection area in the next 1-3 minutes, and obtains the prediction results. A simulation database is built based on all previous prediction results. Step 4, parameter adjustment: Based on the prediction results of the machine learning model, the purification efficiency and energy consumption of the reactor are weighted and allocated. The driving signal of the pulse generation circuit in the reactor is modified according to the weight allocation result, and a safe threshold for ozone concentration is set to control the ozone concentration below the set safe threshold. Step 5, Monitoring and Alarm: Real-time monitoring of the pollution load and purification efficiency of the dust collection area; setting the pollution load threshold for the dust collection area; An alarm message will be sent to staff when one or more of the following situations occur: the pollution load in the dust collection area exceeds the pollution load threshold, or the purification efficiency of the dust collection area decreases by more than 5% within 3-5 minutes under the same pulse parameters.

2. The control method for bipolar dual-reaction zone pulse air purification according to claim 1, characterized in that, In step one, a solid-state circuit breaker is connected in series at the output terminal of the pulse power supply, and a varistor is connected in parallel.

3. The control method for bipolar dual-reaction zone pulse air purification according to claim 2, characterized in that, In step one, the reactor also integrates an IGBT module, and the IGBT module also integrates a current sensor and a temperature sensor; the collector current of the IGBT module is monitored in real time by the current sensor to obtain its current data; the shell temperature of the IGBT module is monitored in real time by the temperature sensor to obtain its temperature data; the current data and temperature data of the IGBT module are input into a machine learning model for analysis to obtain integrated data, and an IGBT aging prediction model is established based on the integrated data. After the staff sets the lifespan threshold of the IGBT module, the IGBT aging prediction model predicts the remaining lifespan of the IGBT module. When the prediction result is lower than the lifespan threshold, an alarm message is sent to the staff.

4. The control method for bipolar dual-reaction zone pulse air purification according to claim 3, characterized in that, In step one, the multi-parameter sensor array includes a PM2.5 concentration sensor, a PM10 concentration sensor, a TVOC concentration sensor, a temperature and humidity sensor, and an ozone concentration sensor, and the sampling frequency of the multi-parameter sensor array is 1-2 seconds / time.

5. The control method for bipolar dual-reaction zone pulse air purification according to claim 4, characterized in that, In step three, the machine learning model updates the simulation database by recording the prediction results after each prediction.

6. The control method for bipolar dual-reaction zone pulse air purification according to claim 5, characterized in that, In step four, when modifying the drive signal of the pulse generation circuit, the waveform used is one of square wave, trapezoidal wave, and resonant waveform.

7. The control method for bipolar dual-reaction zone pulse air purification according to claim 6, characterized in that, In step four, the machine learning model recommends pulse parameters to staff based on the simulation database and adjusts the pulse parameters according to the real-time monitored ozone concentration. When the ozone concentration approaches the safety threshold, the input pulse voltage and frequency are reduced.

8. The control method for bipolar dual-reaction zone pulse air purification according to claim 7, characterized in that, In step four, the pulse parameters include the DC bias voltage V1 of the dust collection zone, and the pulse voltage V2, frequency F2, and pulse width W2 of the pre-ionization zone.

9. The control method for bipolar dual-reaction zone pulse air purification according to claim 8, characterized in that, In step four, when modifying the drive signal of the pulse generation circuit, the modification logic is as follows: When particulate matter is the main pollutant, optimize the pulse parameters of the pre-ionization zone to improve the charging efficiency of particulate matter. When VOCs are the main pollutant type, the pulse parameters of the pre-ionization zone are optimized to enhance the generation of free radicals and the catalytic decomposition of VOCs by non-equilibrium plasma.

10. The control method for bipolar dual-reaction zone pulse air purification according to claim 9, characterized in that, In steps one and five, when sending alarm information to staff, an audible and visual alarm is used, and an alarm SMS is sent to the staff's terminal at the same time.

Images