Automatic membrane replacement method for automatic membrane replacement air filter based on AI

By collecting data from air filters to calculate the filtration quality and efficiency coefficient, and combining it with AI algorithms to control the automatic membrane replacement device, the problem of the air filter membrane's usage being affected by the environment is solved, and timely replacement of the filter membrane and improved air purification effects are achieved.

CN120132519BActive Publication Date: 2025-09-16WUHAN YITE ENVIRONMENTAL PROTECTION TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202510601197.2
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-05-12
Publication Date
2025-09-16
Estimated Expiration
2045-05-12

AI Technical Summary

Technical Problem

Existing AI-based automatic membrane-changing air filters are unable to replace the filter membrane in a timely and accurate manner according to the specific working environment, resulting in poor air purification effects.

Method used

By collecting the standard air flow rate of the air filter, air flow rate, air pressure difference on both sides of the filter membrane and the concentration of air particulate pollutants before and after filtration, the AI ​​algorithm is used to calculate the filtration quality coefficient, filtration effect coefficient and filter membrane efficiency influence coefficient. Combined with historical data, the automatic membrane replacement device is controlled to replace the filter membrane.

Benefits of technology

It can accurately judge whether the filter membrane needs to be replaced according to the use effect of the air filter and environmental changes, thereby improving the purification efficiency of the air filter and the timeliness of the filter membrane use.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120132519B_ABST
    Figure CN120132519B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of air purification technology and proposes an AI-based automatic membrane replacement method for an automatic membrane-changing air filter, comprising: collecting the concentration of air particulate pollutants before and after filtration, air circulation volume, air pressure difference on both sides of the filter membrane, and standard air circulation volume of the air filter at each moment; obtaining the filtration quality coefficient at the target sampling moment and the filtration effect coefficient at the current sampling moment; marking the first mutation moment and the second mutation moment to determine the filter membrane efficiency influence coefficient at the current sampling moment; using AI to control an automatic membrane replacement device to replace the filter membrane based on the filtration effect coefficient and the filter membrane efficiency influence coefficient at the current sampling moment, as well as air filters with different blockage states in historical data. The present invention solves the problem that the use status of the filter membrane of an air filter is affected by the specific working environment, making it impossible to replace the filter membrane in a timely and accurate manner.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of air purification, and in particular to an automatic membrane replacement method for an automatic membrane replacement air filter based on AI. Background Art

[0002] AI-based automatic membrane replacement air filters can automatically replace the membrane based on the membrane's usage time. Generally, the total membrane usage time of the corresponding membrane is one of the membrane's usage parameters. When the air filter's membrane usage time equals the preset total membrane usage time, the AI-controlled automatic membrane replacement device is activated to achieve automatic membrane replacement.

[0003] However, the specific working environments of air filters vary greatly. When an air filter works in an environment with high air pollution, in an environment with a large number of large-particle air pollutants, or in a high-end working state for a long time, the filter membrane will be more damaged, and it is easy to have the problem that the remaining effective use time of the filter membrane is greater than zero but the air purification effect of the air filter is poor. When such a problem occurs, the filter membrane needs to be replaced in time to ensure the use effect of the air filter, but the existing AI-based automatic membrane replacement air filter cannot achieve this. Therefore, there is a need for a method that can solve the problem that the use status of the filter membrane of the air filter is affected by the specific working environment and the filter membrane cannot be replaced in a timely and accurate manner. Summary of the Invention

[0004] The present invention provides an automatic membrane replacement method for an AI-based automatic membrane replacement air filter to solve the problem that the use status of the air filter membrane is affected by the specific working environment and the filter membrane cannot be replaced in a timely and accurate manner. The technical solution adopted is as follows:

[0005] One embodiment of the present invention provides an automatic membrane replacement method for an automatic membrane replacement air filter based on AI, the method comprising the following steps:

[0006] Collect the standard air flow rate of the air filter, the air flow rate of the air filter at the current sampling time and at each moment in the previous local time period, the air pressure difference on both sides of the filter membrane, and the concentration of air particulate pollutants before and after filtration;

[0007] Record any sampling moment as the target sampling moment, and obtain the filtration quality coefficient at the target sampling moment based on the concentration of air particulate pollutants at the air filter outlet after filtration, the difference in air particulate pollutant concentration before and after filtration, and the air circulation volume. Obtain the filtration effect coefficient at the current sampling moment based on the correlation between the filtration quality coefficient and the air circulation volume at each moment in the local time period before the current sampling moment, as well as the data distribution of the filtration quality coefficient;

[0008] Mark the first mutation moment and the second mutation moment according to the change trend of the air pressure difference before and after the current sampling moment, and determine the filter efficiency influence coefficient at the current sampling moment according to the difference between the standard air circulation volume and the air circulation volume at the current sampling moment, and the difference between the air pressure difference at the current sampling moment and the air pressure difference at the first mutation moment and the second mutation moment;

[0009] According to the filtration effect coefficient and filter membrane efficiency influence coefficient at the current sampling moment, as well as the air filters in different blockage states in historical data, AI is used to control the automatic membrane changing device of the air filter to automatically change the membrane.

[0010] Furthermore, the specific method for obtaining the filtering quality coefficient at the target sampling moment is as follows:

[0011] The difference between the concentration of air particulate matter pollutants at the outlet of the air filter after filtration at the target sampling time and the qualified standard of the concentration of particulate matter pollutants in harmful air is recorded as the pollutant concentration difference at the target sampling time;

[0012] The product of the pollutant concentration difference, the air particulate pollutant concentration difference and the air circulation volume at the target sampling time is recorded as the filtration quality coefficient at the target sampling time.

[0013] Furthermore, the specific method for obtaining the filtering effect coefficient at the current sampling moment is:

[0014] Arrange the filter quality coefficients in the order of the corresponding sampling moments to obtain a filter quality coefficient sequence;

[0015] The filtering effect coefficient at the current sampling moment is obtained according to the correlation between the flow rate sequence and the filtering quality coefficient sequence at the current sampling moment and the data distribution of the filtering quality coefficient.

[0016] Furthermore, the method of obtaining the filter effect coefficient at the current sampling moment based on the correlation between the flow rate sequence and the filter quality coefficient sequence at the current sampling moment and the data distribution of the filter quality coefficient includes the following specific methods:

[0017] Arrange the flow rates at the current sampling moment and before according to the order of acquisition time to obtain the flow rate sequence at the current sampling moment, and arrange the filter quality coefficients at the current sampling moment and before according to the order of acquisition time to obtain the filter quality coefficient sequence at the current sampling moment;

[0018] The absolute value of the product of the correlation between the flow sequence and the filter quality coefficient sequence at the current sampling moment and the cumulative sum of all filter quality coefficients included in the filter quality coefficient sequence at the current sampling moment is recorded as the filter effect coefficient at the current sampling moment.

[0019] Furthermore, the specific method of marking the first mutation moment and the second mutation moment is:

[0020] Perform curve fitting on the pressure difference sequence at the current sampling moment to obtain the pressure difference curve at the current sampling moment;

[0021] The earliest sampling time of the two points with the largest curvature on the pressure difference curve at the current sampling time is recorded as the first mutation time of the current sampling time; the latest sampling time of the two points with the largest curvature on the pressure difference curve at the current sampling time is recorded as the second mutation time of the current sampling time.

[0022] Furthermore, the method for obtaining the filter membrane efficiency influence coefficient at the current sampling moment is:

[0023] The difference between the standard air flow rate of the air filter at the current sampling moment and the air flow rate at the current sampling moment is recorded as the first difference at the current sampling moment, and the ratio of the first difference at the current sampling moment to the standard air flow rate of the air filter at the current sampling moment is recorded as the first ratio at the current sampling moment;

[0024] The absolute value of the difference between the pressure difference on both sides of the filter membrane at the current sampling moment and the pressure difference at the first mutation moment is recorded as the first absolute value at the current sampling moment, and the absolute value of the difference between the pressure difference on both sides of the filter membrane at the current sampling moment and the pressure difference at the second mutation moment is recorded as the second absolute value at the current sampling moment;

[0025] The filter membrane efficiency influence coefficient at the current sampling moment is obtained according to the first ratio, the first absolute value and the second absolute value at the current sampling moment.

[0026] Furthermore, the method for obtaining the filter membrane efficiency influence coefficient at the current sampling moment is:

[0027] The sum of the second absolute value at the current sampling moment and the preset parameter is recorded as the first sum value at the current sampling moment, the ratio of the first absolute value at the current sampling moment to the first sum value is recorded as the second ratio at the current sampling moment, and the product of the second ratio at the current sampling moment and the first ratio is the filter membrane efficiency influence coefficient at the current sampling moment.

[0028] Furthermore, the method of using AI to control the automatic membrane replacement device of the air filter based on the filtration effect coefficient and the filter membrane efficiency influence coefficient at the current sampling moment and the air filters in different blockage states in historical data, and automatically replacing the membrane using AI to control the automatic membrane replacement device of the air filter includes the following specific methods:

[0029] Collecting historical data of a preset number of air filters in different blockage states before and after filtration, the concentration of air particulate matter pollutants, air flow rate, and the air pressure difference across the filter membrane, and obtaining the filtration effect coefficient and the filter membrane efficiency influence coefficient at the preset number of historical sampling moments; obtaining a first and a second dividing threshold value based on the filtration effect coefficient and the filter membrane efficiency influence coefficient at the historical sampling moments;

[0030] According to the filtration effect coefficient and filter membrane efficiency influence coefficient at the current sampling moment, as well as the first division threshold and the second division threshold, AI is used to control the automatic membrane changing device of the air filter to automatically change the membrane.

[0031] Furthermore, the method of using AI to control the automatic membrane changing device of the air filter to automatically change the membrane according to the filtration effect coefficient and the filter membrane efficiency influence coefficient at the current sampling moment, as well as the first division threshold and the second division threshold, includes the following specific methods:

[0032] When the filtration effect coefficient at the current sampling moment is less than the first classification threshold, or the filter membrane efficiency influence coefficient at the current sampling moment is greater than the second classification threshold, the AI ​​is used to activate the automatic membrane replacement device of the air filter to replace the filter membrane of the air filter;

[0033] When the filtration effect coefficient at the current sampling moment is greater than or equal to the first classification threshold, and the filter membrane efficiency influence coefficient at the current sampling moment is less than or equal to the second classification threshold, the filter membrane of the air filter is not replaced.

[0034] The beneficial effects of the present invention are:

[0035] The present invention first evaluates the effect of the air filter based on the difference in air particulate pollutant concentration before and after the air is treated by the air filter and the air circulation volume, and evaluates the working environment of the air filter in combination with the air particulate pollutant concentration at the outlet after filtration by the air filter, and obtains the filtration quality coefficient at the target sampling moment. Furthermore, based on the correlation between the filtration quality coefficient and the air circulation volume at each moment in the local time period before the current sampling moment, as well as the value of the filtration quality coefficient, the filtration effect of the filter membrane of the air filter is evaluated to obtain the filtration effect coefficient at the current sampling moment; then, considering that the air circulation volume cannot stably reflect the degree of clogging of the filter membrane when environmental factors change and the filter membrane is physically damaged, the present invention evaluates the filtration efficiency of the air filter in combination with the air pressure difference on both sides of the filter membrane, and marks the first mutation moment and the second mutation moment according to the changing trend of the air pressure difference between the current and previous sampling moments. The first mutation moment is The filter membrane starts to become seriously clogged, and the second mutation moment is the moment when the degree of filter membrane clogs reaches the limit. According to the difference between the standard air circulation volume and the air circulation volume at the current sampling moment, and the difference between the air pressure difference of the air circulation volume and the air pressure difference at the first mutation moment and the second mutation moment, the filtration efficiency of the filter membrane of the air filter is evaluated, and the filter membrane efficiency influence coefficient at the current sampling moment is determined; finally, according to the filtration effect and filtration efficiency evaluation of the filter membrane of the air filter, as well as air filters with different clogging states in historical data, AI is used to control the automatic membrane changing device of the air filter to automatically change the membrane; the present invention analyzes the use effect and use environment of the air filter, and controls the automatic membrane changing device of the air filter according to the filtration effect and filtration efficiency evaluation of the filter membrane of the air filter, so as to solve the problem that the use status of the filter membrane of the air filter is affected by the specific working environment and the filter membrane cannot be replaced in time and accurately. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 A schematic flow chart of an automatic membrane replacement method for an AI-based automatic membrane replacement air filter provided in one embodiment of the present invention;

[0038] Figure 2 A schematic diagram of an air pressure difference curve provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0039] 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.

[0040] See also Figure 1 , which shows a flow chart of an automatic membrane replacement method for an automatic membrane replacement air filter based on AI provided by one embodiment of the present invention, the method comprising the following steps:

[0041] Step S001, collect the standard air flow rate of the air filter, the air flow rate of the air filter at the current sampling moment and at each moment in the previous local time period, the air pressure difference on both sides of the filter membrane, and the air particulate pollutant concentration before and after filtration.

[0042] An automatic membrane-changing air filter is one in which an automatic membrane-changing device is installed in the filter membrane position of the air filter. The automatic membrane-changing device automatically changes the membrane of the air filter. The automatic membrane-changing device of the air filter consists of a material barrel, filter material, material push head, linear module, electric push rod, spring device, discharge port, pressure plate and air pipe connector.

[0043] Monitoring equipment inside the air filter collects the concentration of particulate matter at the filter inlet and at the filter outlet after filtration, as well as the air flow rate. Pressure sensors on both sides of the membrane in the dynamic membrane exchange device collect the air pressure across the membrane and calculate the pressure differential across the membrane. Based on the factory parameters of the air filter, the standard air flow rate per hour for each filter setting is calculated.

[0044] The difference in air particulate matter concentration before and after filtration of the air filter at the same sampling time is recorded as the air particulate matter concentration difference. The difference in air particulate matter concentration before and after filtration of the air filter at the same sampling time is the difference in air particulate matter concentration at the inlet and outlet of the air filter at the same sampling time.

[0045] Arrange the flow rates before and after the current sampling moment in the order of collection time to obtain the flow rate sequence at the current sampling moment; arrange the pressure differences before and after the current sampling moment in the order of collection time to obtain the pressure difference sequence at the current sampling moment.

[0046] Among them, the air pressure difference on both sides of the filter membrane is the maximum value minus the minimum value of the air pressure on both sides of the filter membrane, so the air pressure difference on both sides of the filter membrane is a non-negative number; in the process of collecting data, this embodiment sets the collection time interval of air particulate pollutant concentration, air circulation and air pressure difference to one hour, and sets the length of the circulation sequence and the air pressure difference sequence to 168, that is, each circulation sequence contains one week's air circulation, and each air pressure difference sequence contains one week's air pressure difference. As other implementation methods, the implementer can decide the collection time interval of all data and the length of all sequences according to actual conditions, and this application does not impose any special restrictions.

[0047] At this point, the standard air flow rate, air particulate pollutant concentration difference, concentration difference sequence, flow rate sequence and air pressure difference sequence are obtained.

[0048] Step S002: record any sampling moment as the target sampling moment, and obtain the filtration quality coefficient of the target sampling moment based on the air particulate pollutant concentration at the outlet of the air filter after filtration at the target sampling moment, the difference in air particulate pollutant concentration before and after filtration by the air filter, and the air circulation volume. Obtain the filtration effect coefficient of the current sampling moment based on the correlation between the filtration quality coefficient and the air circulation volume at each moment in the local time period before the current sampling moment, as well as the data distribution of the filtration quality coefficient.

[0049] When the remaining effective use time of the filter membrane of the air filter decreases and the filtering quality of the air filter drops to a certain level, it is necessary to consider replacing the filter membrane. Generally, the remaining effective use time of the filter membrane of the air filter can be obtained by subtracting the use time of the filter membrane from the preset total use time of the filter membrane. When the remaining effective use time of the filter membrane is equal to zero, the filter membrane is replaced. However, the specific working environment of the air filter varies greatly. When the air filter works in an environment with high air pollution, in an environment with a large number of large-particle air pollutants, or is in a high-end working state for a long time, the loss of the filter membrane is greater, and it is easy to have the problem that the remaining effective use time of the filter membrane is greater than zero but the air purification effect of the air filter is poor. When such a problem occurs, the filter membrane also needs to be replaced in time. Therefore, it is necessary to comprehensively consider the use environment, use time and filtering effect of the air filter to determine whether the filter membrane of the air filter needs to be replaced.

[0050] Any sampling moment is recorded as the target sampling moment. This embodiment is described by taking any sampling moment as an example.

[0051] A filtration quality coefficient at the target sampling time is obtained based on the air particulate matter concentration at the air filter outlet after filtration, the air particulate matter concentration difference, and the air circulation volume at the target sampling time. The filtration quality coefficient at the target sampling time is positively correlated with the air particulate matter concentration difference and the air circulation volume at the target sampling time, and is negatively correlated with the air particulate matter concentration at the air filter outlet after filtration at the target sampling time.

[0052] Preferably, as an embodiment of the present application, the difference between the concentration of air particulate pollutants at the outlet of the air filter after filtration at the target sampling time and the qualified standard of the concentration of particulate pollutants in harmful air is recorded as the pollutant concentration difference at the target sampling time; the product of the pollutant concentration difference at the target sampling time, the air particulate pollutant concentration difference and the air circulation volume is recorded as the filtration quality coefficient at the target sampling time. Among them, the qualified standard value of the concentration of particulate pollutants in harmful air in this embodiment is 200 The implementer can decide the value of the qualified standard for the concentration of particulate matter pollutants in harmful air based on actual conditions, and this application does not impose any special restrictions.

[0053] When the filter membrane of the air filter does not need to be replaced, the air filter has a better performance. The difference between the concentration of air particulate pollutants at the outlet of the air filter after filtration and the qualified standard of the concentration of particulate pollutants in harmful air is small. The difference in air particulate pollutant concentration and air circulation at the target sampling time are large. The air filter can filter out more air particulates. At this time, the filtration quality coefficient at the target sampling time is large.

[0054] The filter quality coefficient at each sampling moment can be obtained in the same way. That is, each sampling moment has a corresponding filter quality coefficient.

[0055] Arrange the filter quality coefficients at the current sampling moment and before in the order of acquisition time to obtain the filter quality coefficient sequence at the current sampling moment. The filter quality coefficient sequence has the same length as the flow rate sequence.

[0056] The filtering effect coefficient at the current sampling moment is obtained according to the correlation between the flow sequence and the filtering quality coefficient sequence at the current sampling moment and the value of the filtering quality coefficient.

[0057] Preferably, as an embodiment of the present application, the absolute value of the product of the correlation between the flow rate sequence and the filter quality coefficient sequence at the current sampling moment and the cumulative sum of all filter quality coefficients contained in the filter quality coefficient sequence at the current sampling moment is recorded as the filter effect coefficient at the current sampling moment.

[0058] Among them, the calculation formula of the filtering effect coefficient at the current sampling moment is:

[0059]

[0060] in, The current sampling time The filtering effect coefficient; The current sampling time The cumulative sum of all filter quality coefficients contained in the filter quality coefficient sequence; The current sampling time The correlation between the flux series and the filter quality coefficient series.

[0061] Among them, this embodiment selects cosine similarity as the correlation measurement between the circulation sequence and the filter quality coefficient sequence. The implementer can use other existing methods such as the Pearson correlation coefficient to measure the correlation between the circulation sequence and the filter quality coefficient sequence. This application does not impose any special restrictions.

[0062] The Filtration Effectiveness Coefficient (FEC) evaluates the filtration efficiency of an air filter's membrane. When the membrane doesn't need to be replaced, the air filter performs well, the Filtration Quality Coefficient (FQC) is high, and the cumulative sum of all the FQCs within the FQC sequence is also high. In this case, the FEC is high. When the membrane doesn't need to be replaced, the air filter performs well, and the correlation between the air flow rate and the FQC is strong. That is, the correlation between the air flow rate sequence and the FQC sequence is strong, and in this case, the FEC is high.

[0063] At this point, the filtering effect coefficient at the current sampling moment is obtained.

[0064] Step S003, marking the first mutation moment and the second mutation moment according to the changing trend of the air pressure difference before and after the current sampling moment, and determining the filter membrane efficiency influence coefficient at the current sampling moment according to the difference between the standard air circulation volume and the air circulation volume at the current sampling moment, and the difference between the air pressure difference at the current sampling moment and the air pressure difference at the first mutation moment and the second mutation moment.

[0065] Generally, in the filtration efficiency evaluation process of existing air filters, the filtration efficiency is evaluated based on the difference between the air circulation volume at each sampling moment and the standard air circulation volume when the air filter is operating at the sampling moment. When the filtration efficiency of the filter membrane of the air filter is low, it is unable to filter particulate matter in the air in a timely manner. In this case, the filter membrane needs to be replaced. However, the air circulation volume may be affected by various environmental factors such as temperature and humidity. When these environmental factors change, the air circulation volume cannot stably reflect the degree of filter membrane clogging. When the filter membrane is physically damaged, such as broken or deformed, the air circulation volume also cannot stably reflect the degree of filter membrane clogging. Therefore, this embodiment evaluates the filtration efficiency of the air filter based on the air pressure difference on both sides of the filter membrane.

[0066] During the use of an air filter, the filter membrane continuously filters particulate matter in the air, and the filter membrane will gradually become clogged. As a result, the air flow rate will continue to decrease, and the pressure difference across the filter membrane will gradually increase. Specifically, during the use of the air filter, in the initial stage, the air flow rate and the pressure difference across the filter membrane do not change significantly. As the degree of filter blockage becomes more and more serious, from a certain point on, the pressure difference across the filter membrane will suddenly increase sharply. The moment when the pressure difference suddenly increases sharply is the moment when the filter membrane becomes seriously clogged. Then, with the continued use of the air filter, the degree of filter blockage gradually increases until the degree of filter blockage reaches the limit, and the pressure difference across the filter membrane begins to slowly increase.

[0067] Using the sampling time as the independent variable and the pressure difference across the filter membrane as the dependent variable, a curve fitting is performed on the pressure difference sequence at the current sampling time to obtain the pressure difference curve at the current sampling time. The two points with the largest curvature on the pressure difference curve at the current sampling time are obtained. The earliest sampling time of the two points with the largest curvature is recorded as the first mutation time of the current sampling time, and the latest sampling time of the two points with the largest curvature is recorded as the second mutation time of the current sampling time. The first mutation time is when the filter membrane becomes seriously clogged, and the second mutation time is when the filter membrane clogs to the extreme.

[0068] The schematic diagram of the pressure difference curve is as follows Figure 2 As shown in the pressure difference curve diagram, the horizontal axis is the sampling time, the vertical axis is the pressure difference, 1 is the first mutation time, and 2 is the second mutation time. In the pressure difference curve, before the first mutation time, the air filter membrane is considered to have almost no clogging problem; at the first mutation time, the membrane clogging becomes severe; between the first mutation time and the second mutation time, the severity of the membrane clogging gradually increases; at the second mutation time, the membrane clogging reaches its limit; after the second mutation time, the membrane is completely clogged.

[0069] Among them, this embodiment uses the least squares method to perform curve fitting on the pressure difference sequence. As other implementation methods, on the basis of achieving the purpose of curve fitting on the pressure difference sequence, the implementer can use other methods of the existing technology to perform curve fitting on the pressure difference sequence. This application does not impose any special restrictions. Using the least squares method to perform curve fitting on the pressure difference sequence and calculate the curvature on the curve is a well-known technology and will not be repeated here.

[0070] The filter membrane efficiency influence coefficient at the current sampling moment is determined based on the difference between the standard air circulation volume and the air circulation volume at the current sampling moment, and the difference between the air pressure difference of the air circulation volume and the air pressure difference at the first mutation moment and the second mutation moment. The filter membrane efficiency influence coefficient at the current sampling moment is positively correlated with the difference between the standard air circulation volume and the air circulation volume at the current sampling moment, positively correlated with the difference between the air pressure difference of the air circulation volume and the air pressure difference at the first mutation moment, and negatively correlated with the difference between the air pressure difference of the air circulation volume and the air pressure difference at the second mutation moment.

[0071] Preferably, as an embodiment of the present application, the difference between the standard air flow rate of the air filter at the current sampling moment and the air flow rate at the current sampling moment is recorded as the first difference at the current sampling moment, and the ratio of the first difference at the current sampling moment to the standard air flow rate of the air filter at the current sampling moment is recorded as the first ratio at the current sampling moment. The absolute value of the difference between the air pressure difference on both sides of the filter membrane at the current sampling moment and the air pressure difference at the first mutation moment is recorded as the first absolute value at the current sampling moment, and the absolute value of the difference between the air pressure difference on both sides of the filter membrane at the current sampling moment and the air pressure difference at the second mutation moment is recorded as the second absolute value at the current sampling moment; the sum of the second absolute value at the current sampling moment and the preset parameter is recorded as the first sum value at the current sampling moment, and the ratio of the first absolute value at the current sampling moment to the first sum value is recorded as the second ratio at the current sampling moment. The product of the second ratio at the current sampling moment and the first ratio is the filter membrane efficiency influence coefficient at the current sampling moment.

[0072] Specifically, the calculation formula for the filter membrane efficiency influence coefficient at the current sampling moment is:

[0073]

[0074] in, The current sampling time The influence coefficient of filter membrane efficiency; The air filter at the current sampling time Standard air flow rate; The current sampling time Air circulation; The current sampling time The absolute value of the difference between the pressure difference on both sides of the filter membrane and the pressure difference at the first mutation moment; The current sampling time The absolute value of the difference between the pressure difference on both sides of the filter membrane and the pressure difference at the second mutation moment; is a preset parameter. In this embodiment, the parameter The value of is 0.01, the parameter The purpose is to prevent the denominator from being 0.

[0075] The filter membrane efficiency coefficient is an evaluation of the filter efficiency of the air filter membrane. When the difference between the standard air flow rate and the air flow rate at the current sampling moment is greater, the amount of air that the air filter membrane can process and pass through is less, and the filter membrane is more seriously clogged. At this time, the filter membrane efficiency coefficient at the current sampling moment is greater. When the difference between the pressure difference of the air flow rate and the pressure difference at the first mutation moment is greater than the difference between the pressure difference of the air flow rate and the pressure difference at the second mutation moment, that is, The larger it is, the closer the air pressure difference at the current sampling moment is to the air pressure difference at the second mutation moment, and the more serious the filter membrane is blocked. At this time, the filter membrane efficiency influence coefficient at the current sampling moment is greater.

[0076] At this point, the filter membrane efficiency influence coefficient at the current sampling moment is obtained.

[0077] Step S004: Use AI to control the automatic membrane changing device of the air filter to automatically change the membrane according to the filtration effect coefficient and the filter membrane efficiency influence coefficient at the current sampling moment, as well as the air filters in different blockage states in the historical data.

[0078] Collect historical data for a preset number of air filters with varying blockage states before and after filtration, including particulate matter concentration, air flow rate, and pressure differential across the filter membrane. Obtain the filtration effectiveness coefficient and filter membrane efficiency influence coefficient for a preset number of historical sampling moments based on the historical data, using the same method used to obtain the filter membrane efficiency influence coefficient at the current sampling moment.

[0079] Among them, this embodiment selects 300 as the preset number, and collects two filter membrane states in total, and the two filter membrane states are respectively that the filter membrane needs to be replaced and that the filter membrane does not need to be replaced. The number of filter membrane efficiency influence coefficients and filtration effect coefficients at the historical sampling moments when the filter membrane needs to be replaced is 200, and the number of filter membrane efficiency influence coefficients and filtration effect coefficients at the historical sampling moments when the filter membrane does not need to be replaced is 100. As other implementation methods, the implementer can decide the number of filter membrane efficiency influence coefficients and filtration effect coefficients at the historical sampling moments corresponding to the two filter membrane states according to actual conditions, and this application does not impose any special restrictions.

[0080] The Otsu threshold segmentation algorithm is used for the filter effect coefficients of all historical sampling moments to obtain the first partition threshold; the Otsu threshold segmentation algorithm is used for the filter membrane efficiency influence coefficients of all historical sampling moments to obtain the second partition threshold.

[0081] Among them, this embodiment selects the Otsu threshold segmentation algorithm to obtain the first division threshold according to the filtering effect coefficient and the second division threshold according to the filter membrane efficiency influence coefficient. On the basis of the purpose of obtaining the first division threshold and the second division threshold, the implementer can use other existing technology methods such as K-means to obtain the first division threshold and the second division threshold. This application does not impose any special restrictions; using the Otsu threshold segmentation algorithm to obtain the first division threshold and the second division threshold is a well-known technology and will not be repeated.

[0082] When the filtration effect coefficient at the current sampling moment is less than the first division threshold, or the filter membrane efficiency influence coefficient at the current sampling moment is greater than the second division threshold, it is judged that the filter membrane of the air filter needs to be replaced, and AI is used to start the automatic membrane changing device of the air filter to replace the filter membrane of the air filter; when the filtration effect coefficient at the current sampling moment is greater than or equal to the first division threshold, and the filter membrane efficiency influence coefficient at the current sampling moment is less than or equal to the second division threshold, it is judged that the filter membrane of the air filter does not need to be replaced.

[0083] At this point, automatic membrane replacement of air filters based on AI is realized.

[0084] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. The automatic membrane replacement method of the automatic membrane replacement air filter based on AI is characterized in that: The method comprises the following steps: Collect the standard air flow rate of the air filter, the air flow rate of the air filter at the current sampling time and at each moment in the previous local time period, the air pressure difference on both sides of the filter membrane, and the concentration of air particulate pollutants before and after filtration; Record any sampling moment as the target sampling moment, record the difference between the concentration of air particulate pollutants at the outlet of the air filter after filtration at the target sampling moment and the qualified standard of the concentration of particulate pollutants in the harmful air as the pollutant concentration difference at the target sampling moment; record the product of the pollutant concentration difference at the target sampling moment, the air particulate pollutant concentration difference and the air circulation volume as the filtration quality coefficient at the target sampling moment; arrange the filtration quality coefficients in the order of the corresponding sampling moments to obtain a filtration quality coefficient sequence, arrange the circulation volumes at the current sampling moment and before in the order of the collection time to obtain the circulation volume sequence at the current sampling moment, arrange the filtration quality coefficients at the current sampling moment and before in the order of the collection time to obtain the filtration quality coefficient sequence at the current sampling moment, and record the absolute value of the product of the correlation between the circulation volume sequence at the current sampling moment and the filtration quality coefficient sequence and the cumulative sum of all filtration quality coefficients contained in the filtration quality coefficient sequence at the current sampling moment as the filtration effect coefficient at the current sampling moment; Perform curve fitting on the pressure difference sequence at the current sampling moment to obtain the pressure difference curve at the current sampling moment; record the earliest sampling moment of the two points with the largest curvature on the pressure difference curve at the current sampling moment as the first mutation moment of the current sampling moment; record the latest sampling moment of the two points with the largest curvature on the pressure difference curve at the current sampling moment as the second mutation moment of the current sampling moment; Record the difference between the standard air circulation volume of the air filter at the current sampling moment and the air circulation volume at the current sampling moment as the first difference at the current sampling moment, and record the ratio of the first difference at the current sampling moment to the standard air circulation volume of the air filter at the current sampling moment as the first ratio at the current sampling moment; record the absolute value of the difference between the air pressure difference on both sides of the filter membrane at the current sampling moment and the air pressure difference at the first mutation moment as the first absolute value at the current sampling moment, and record the absolute value of the difference between the air pressure difference on both sides of the filter membrane at the current sampling moment and the air pressure difference at the second mutation moment as the second absolute value at the current sampling moment; obtain the filter membrane efficiency influence coefficient at the current sampling moment based on the first ratio, the first absolute value, and the second absolute value at the current sampling moment; According to the filtration effect coefficient and filter membrane efficiency influence coefficient at the current sampling moment, as well as the air filters in different blockage states in historical data, AI is used to control the automatic membrane changing device of the air filter to automatically change the membrane.

2. The automatic membrane replacement method of the automatic membrane replacement air filter based on AI according to claim 1 is characterized in that: The method for obtaining the filter membrane efficiency influence coefficient at the current sampling moment is: The sum of the second absolute value at the current sampling moment and the preset parameter is recorded as the first sum value at the current sampling moment, the ratio of the first absolute value at the current sampling moment to the first sum value is recorded as the second ratio at the current sampling moment, and the product of the second ratio at the current sampling moment and the first ratio is the filter membrane efficiency influence coefficient at the current sampling moment.

3. The automatic membrane replacement method of the automatic membrane replacement air filter based on AI according to claim 1 is characterized in that: The method of using AI to control the automatic membrane changing device of the air filter based on the filtration effect coefficient and the filter membrane efficiency influence coefficient at the current sampling moment and the air filters in different blockage states in historical data, and using AI to control the automatic membrane changing device of the air filter to automatically change the membrane includes the following specific methods: Collecting historical data of a preset number of air filters in different blockage states before and after filtration, the concentration of air particulate matter pollutants, air flow rate, and the air pressure difference across the filter membrane, and obtaining the filtration effect coefficient and the filter membrane efficiency influence coefficient at the preset number of historical sampling moments; obtaining a first and a second dividing threshold value based on the filtration effect coefficient and the filter membrane efficiency influence coefficient at the historical sampling moments; According to the filtration effect coefficient and filter membrane efficiency influence coefficient at the current sampling moment, as well as the first division threshold and the second division threshold, AI is used to control the automatic membrane changing device of the air filter to automatically change the membrane.

4. The automatic membrane replacement method of the automatic membrane replacement air filter based on AI according to claim 3 is characterized in that: The specific method of using AI to control the automatic membrane changing device of the air filter to automatically change the membrane according to the filtration effect coefficient and the filter membrane efficiency influence coefficient at the current sampling moment, as well as the first division threshold and the second division threshold, includes: When the filtration effect coefficient at the current sampling moment is less than the first classification threshold, or the filter membrane efficiency influence coefficient at the current sampling moment is greater than the second classification threshold, the AI ​​is used to activate the automatic membrane replacement device of the air filter to replace the filter membrane of the air filter; When the filtration effect coefficient at the current sampling moment is greater than or equal to the first classification threshold, and the filter membrane efficiency influence coefficient at the current sampling moment is less than or equal to the second classification threshold, the filter membrane of the air filter is not replaced.

Citation Information

Patent Citations

  • Sampler with automatically replaced filter membrane and adsorption device and application method thereof

    CN103900862A

  • Fresh air handling unit filter screen replacement method, device and system and storage medium

    CN119713496A