A method and system for monitoring air filter IoT smart chips

Abstract

The present invention provides an Internet of Things (IoT) intelligent chip monitoring method and system for air filter elements, relating to the field of intelligent management technology. The method comprises: using a laser particulate matter sensor to collect the PM2.5 or PM10 concentration and dust holding capacity of the filter element; using a differential pressure sensor to monitor the pressure differential across the filter element, and using a temperature and humidity compensation algorithm to correct the pressure differential data; using a non-dispersive infrared carbon dioxide sensor to detect intake air concentration and identify air-fuel ratio anomalies with sudden changes in carbon dioxide concentration greater than 15%; extracting eigenvalues ​​based on the dust holding capacity and laser scattering intensity matrix eigenvalues ​​and matching them with a preset dust feature library to output dust type labels and wear coefficients; and inputting air permeability time series, ambient temperature and humidity, dust load, and engine operating conditions based on the dust holding capacity, corrected differential pressure data, ambient temperature and humidity, and engine operating conditions to calculate the remaining filter life percentage, with a prediction error of less than ±8%. The present invention enables intelligent management of filter elements throughout their entire life cycle.

Description

Technical Field

[0001] This invention relates to the field of intelligent management and processing technology, specifically to a method and system for monitoring an air filter using an IoT intelligent chip. Background Technology

[0002] Traditional air filter monitoring technology has several drawbacks. It relies on mechanical differential pressure sensors or fixed-cycle replacement strategies, making it susceptible to interference from dust type and temperature / humidity. In high-humidity environments, dust agglomeration reduces the sensitivity of the differential pressure sensor, failing to accurately reflect the filter's clogging status. Furthermore, traditional devices are not integrated with the vehicle's ECU or cloud platform, only providing local alarms that require manual recording by the driver and contacting a service station, leading to delays in maintenance response. Additionally, the repetitive sensor reset and system data update operations during filter replacement further increase the risk of human error.

[0003] In terms of filter element lifecycle management, most existing filter elements are designed for single use and lack chip-level identification and lifespan tracking functions. In actual use, filter elements usually need to undergo multiple maintenance before they are scrapped. As the number of maintenance increases, their air permeability decreases and the material gradually deteriorates until they reach the replacement and scrapping standard. However, traditional technology cannot accurately track this dynamic change process, resulting in 30% of the remaining lifespan of scrapped filter elements in a certain mining area not being effectively utilized, causing resource waste.

[0004] In terms of user interaction, drivers need to check the in-vehicle dashboard, mobile app, and paper maintenance manual simultaneously. The lack of integration of information from multiple platforms causes some drivers to ignore warnings due to the complexity of operation, ultimately increasing vehicle maintenance costs. In terms of technical functions, traditional solutions also lack intelligent functions such as dust composition analysis, dynamic prediction of remaining lifespan, and visualization of carbon emission savings. Overall, they are limited to the collection of single physical signals and have not formed a closed loop of "perception-analysis-decision-service", resulting in high user maintenance costs, serious waste of resources, and failure to effectively release the value of data. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an IoT smart chip monitoring method and system for air filter elements, so as to realize intelligent management of the entire life cycle of the filter element.

[0006] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0007] In a first aspect, a method for monitoring air filter IoT smart chips, the method comprising:

[0008] Step 1: The PM2.5 or PM10 concentration and dust holding capacity of the filter element are collected at 10-second intervals using a laser particulate sensor. The pressure difference before and after the filter element is monitored by a differential pressure sensor, and the pressure difference data is corrected using a temperature and humidity compensation algorithm. The intake air concentration is detected by a non-dispersive infrared carbon dioxide sensor to identify abnormal air-fuel ratios where the carbon dioxide concentration changes by more than 15%.

[0009] Step 2: Based on the dust holding capacity and the laser scattering intensity matrix feature values, extract the feature values ​​and match them with the preset dust feature library, output the dust type label and wear coefficient, and input the air permeability time series, ambient temperature and humidity, dust load and engine operating conditions according to the dust holding capacity, corrected differential pressure data, ambient temperature and humidity, and engine operating conditions to obtain the remaining life percentage of the filter element, with a prediction error of <±8%;

[0010] Step 3: Based on the output remaining lifespan percentage and air permeability threshold, trigger a level 3 warning:

[0011] A pop-up notification will be executed when the air permeability is ≤50%.

[0012] When the air permeability is ≤40%, a pop-up window and SMS notification will be executed.

[0013] When the air permeability is ≤20%, a pop-up reminder will be displayed on the screen and the fuel consumption increment will be calculated.

[0014] Step 4: When the remaining lifespan of the output is <5%, the filter element location data is sent to the cloud. The cloud generates a reverse logistics work order. The company arranges after-sales personnel to visit the driver and recommend that the driver perform filter element maintenance. For filter elements with a residual value assessment >30%, the return-to-factory remanufacturing process is initiated.

[0015] Furthermore, the PM2.5 or PM10 concentration and dust holding capacity of the filter element are collected at 10-second intervals using a laser particulate sensor, including:

[0016] Step 1.1: Using a MEMS laser scattering sensor, synchronously collect PM2.5 concentration, PM10 concentration, and dust holding capacity data at a fixed interval of 10 seconds to obtain the collected data;

[0017] Step 1.2: Based on the collected data, outliers caused by momentary environmental interference are removed, the dust holding capacity is converted into dust load, and the verified PM2.5 or PM10 concentration and dust load are encrypted and uploaded to the cloud through a low-power communication module.

[0018] Furthermore, a differential pressure sensor monitors the pressure difference across the filter element, and a temperature and humidity compensation algorithm is used to correct the pressure difference data. A non-spectral infrared carbon dioxide sensor detects the intake air concentration and identifies air-fuel ratio anomalies where the carbon dioxide concentration surges by more than 15%, including:

[0019] Step 1.3: The temperature and relative humidity of the environment where the filter element is located are obtained in real time by the temperature and humidity sensor integrated on the chip. The original output value of the differential pressure sensor is input into the linear compensation formula, correlated with the dust load data, and high clogging risk areas are marked.

[0020] Step 1.4: Based on high congestion risk areas, when a sudden change in carbon dioxide concentration of a single vehicle is detected to be >15%, retrieve carbon dioxide data of other vehicles within 10 kilometers of the same area. If the proportion of vehicles with sudden changes in the area is >30%, it is determined to be an environmental factor. If only a single vehicle has a sudden change, correlate it with the engine operating data of that vehicle to determine a fuel injection failure.

[0021] Furthermore, based on the dust holding capacity and the eigenvalues ​​of the laser scattering intensity matrix, eigenvalues ​​are extracted and matched against a pre-defined dust feature library, outputting dust type labels and wear coefficients, including:

[0022] Step 2.1: Based on the corrected differential pressure data and dust load, extract the feature values ​​of the laser scattering intensity matrix, including calculating the average, variance, and peak value of the scattering intensity as feature vectors; perform similarity matching between the feature vectors and the preset dust feature library, and output dust type labels; simultaneously, fuse differential pressure-dust coupled heat map data to mark the distribution areas of high-wear dust; when more than 50 vehicles in the same geographical area identify a new dust type that is not preset, automatically expand the categories of the dust feature library and send it to the edge computing unit for updating;

[0023] Step 2.2: Based on the dust type label, connect to the vehicle manufacturer's ECU historical data platform to retrieve the engine cylinder pressure fluctuation value during the filter's service life; calculate the wear coefficient based on the cylinder pressure fluctuation value, specifically by statistically analyzing the standard deviation of the cylinder pressure fluctuation to quantify the degree of wear, and output the wear coefficient in association with the dust type label.

[0024] Furthermore, based on dust holding capacity, corrected differential pressure data, ambient temperature and humidity, and engine operating conditions, the permeability time series, ambient temperature and humidity, dust load, and engine operating conditions are input to obtain the remaining lifespan percentage of the filter element. The prediction error is <±8%, including:

[0025] Step 2.3: Match the preset air permeability decay baseline curve according to the dust type label; adjust the slope of the decay curve by superimposing the wear coefficient; fuse the corrected differential pressure data and dust load to calculate the real-time air permeability decay value to obtain the initial value of remaining life; perform weighted fusion calculation with the regional working condition feature library and historical filter scrapping data, where the regional working condition feature weight accounts for 60% and the historical data weight accounts for 40%, and output the optimized value of remaining life percentage with a prediction error of <±8%;

[0026] Step 2.4: When the filter element is actually scrapped, dynamic compensation is performed based on the optimized remaining lifespan percentage output in Step 5.1.

[0027] The ratio of the predicted lifespan in days to the actual number of days used is calculated as an error compensation factor.

[0028] If the error compensation factor is greater than 1.1, add the attenuation rate compensation value for high dust conditions to the air permeability attenuation baseline curve.

[0029] If the error compensation factor is less than 0.9, a dust environment correction coefficient of 0.15 is added to the dust load.

[0030] Furthermore, based on the output remaining lifespan percentage and air permeability threshold, a three-level warning is triggered: when air permeability ≤ 50%, a pop-up notification is executed; when air permeability ≤ 40%, both pop-up and SMS notifications are executed; when air permeability ≤ 20%, a screen pop-up reminder is executed and fuel consumption increment is calculated, including:

[0031] Step 3.1: Read the breathability decay value corresponding to the remaining lifespan percentage optimization value in real time. When the breathability decay value reaches the initial threshold of 50%, activate the IoT communication module to send an APP pop-up and SMS notification to the user terminal.

[0032] Step 3.2: When the air permeability decay value continues to drop to 40%, the vehicle voice module is invoked to broadcast the warning content. At the same time, the current pressure difference increase ratio is calculated based on the corrected pressure difference data. If the pressure difference increase ratio is >30%, the warning level is simultaneously upgraded to level two.

[0033] Step 3.3: When the air permeability decay value reaches the critical threshold of 20%, a forced pop-up reminder is executed on the screen, and fuel consumption increment calculation is performed simultaneously.

[0034] Retrieve dust load data and output wear coefficient to calculate dust wear equivalent value;

[0035] Based on the data of the marked high-clogging risk areas, extract the actual value of the pressure difference before and after the current filter element;

[0036] By combining engine operating data, the actual pressure difference value is converted into the engine intake drag coefficient;

[0037] The absolute value of the fuel consumption increase per 100 kilometers is output by mapping the intake drag coefficient to the fuel consumption of the car manufacturer's ECU.

[0038] Furthermore, when the remaining lifespan of the output filter is less than 5%, the filter location data is sent to the cloud. The cloud generates a reverse logistics work order, and the company arranges after-sales personnel to follow up with the driver, advising the driver to perform filter maintenance. For filter elements with a residual value assessment greater than 30%, a return-to-factory remanufacturing process is initiated, including:

[0039] Step 4.1: When the output remaining life percentage optimization value is lower than 5% for the first time, retrieve the real-time location data of the vehicle's GPS module, combine it with the filter serial number to generate a unique identifier, and upload the identifier and location data to the cloud server through the IoT communication module.

[0040] Step 4.2: Based on the received filter cartridge serial number, the company arranges after-sales personnel to follow up with the driver, recommending filter cartridge maintenance, and retrieving the output wear coefficient and recorded dynamic compensation historical data to perform residual value assessment calculation.

[0041] Extract the total cumulative dust load of the filter element and divide it by the preset maximum dust holding capacity threshold of the filter material to obtain the physical loss rate. Weight the wear coefficient and the physical loss rate, with weights of 70% and 30% respectively, and output the residual value percentage. If the residual value percentage is greater than 30%, automatically generate a reverse logistics work order containing filter element positioning information and simultaneously perform a remanufacturing feasibility assessment.

[0042] The calculated dust wear equivalent value is retrieved. If the wear equivalent value is ≤0.3 and the error compensation factor is in the range of [0.95, 1.05], a filter element disassembly instruction is sent to the remanufacturing plant. If the wear equivalent value is >0.3, the filter element scrapping process is triggered and the user is notified.

[0043] Secondly, an air filter IoT smart chip monitoring system includes:

[0044] The acquisition module is used to collect the PM2.5 or PM10 concentration and dust holding capacity of the filter element at 10-second intervals using a laser particulate sensor, monitor the pressure difference across the filter element using a differential pressure sensor, and correct the pressure difference data using a temperature and humidity compensation algorithm. It also detects the intake air concentration using a non-dispersive infrared carbon dioxide sensor and identifies abnormal air-fuel ratios where the carbon dioxide concentration changes by more than 15%.

[0045] The extraction module is used to extract feature values ​​and match them with a preset dust feature library based on dust holding capacity and laser scattering intensity matrix feature values, and output dust type labels and wear coefficients. Based on dust holding capacity, corrected differential pressure data, ambient temperature and humidity, and engine operating conditions, the module inputs air permeability time series, ambient temperature and humidity, dust load, and engine operating conditions to obtain the remaining life percentage of the filter element, with a prediction error of <±8%.

[0046] The calculation module is used to trigger a three-level warning based on the output remaining lifespan percentage and air permeability threshold: when the air permeability is ≤50%, a pop-up notification is executed; when the air permeability is ≤40%, a pop-up and SMS notification are executed; when the air permeability is ≤20%, a screen pop-up reminder is executed and the fuel consumption increment is calculated.

[0047] The processing module sends filter location data to the cloud when the remaining lifespan of the output is less than 5%. The cloud generates a reverse logistics work order, and the company arranges after-sales personnel to visit the driver and recommend that the driver perform filter maintenance. For filters with a residual value assessment of more than 30%, the return-to-factory remanufacturing process is initiated.

[0048] Thirdly, a computing device, comprising:

[0049] one or more processors;

[0050] A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0051] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0052] The above-described solution of the present invention has at least the following beneficial effects:

[0053] By using multi-sensor fusion and edge computing technology, real-time and accurate monitoring of air filter status and intelligent management throughout its entire lifecycle are achieved. At the same time, through dynamic early warning mechanisms and circular economy design, excessive replacement and engine damage are avoided, reducing user maintenance costs. Attached Figure Description

[0054] Figure 1 This is a flowchart illustrating an air filter IoT smart chip monitoring method provided by an embodiment of the present invention.

[0055] Figure 2 This is a schematic diagram of an air filter IoT smart chip monitoring system provided by an embodiment of the present invention. Detailed Implementation

[0056] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0057] like Figure 1 As shown, an embodiment of the present invention proposes a method for monitoring air filter IoT smart chips, the method comprising the following steps:

[0058] Step 1: The PM2.5 or PM10 concentration and dust holding capacity of the filter element are collected at 10-second intervals using a laser particulate sensor. The pressure difference before and after the filter element is monitored by a differential pressure sensor, and the pressure difference data is corrected by a temperature and humidity compensation algorithm. The intake air concentration is detected by a non-dispersive infrared carbon dioxide sensor to identify abnormal air-fuel ratios where the carbon dioxide concentration changes by more than 15%.

[0059] Step 2: Based on the dust holding capacity and the laser scattering intensity matrix feature values, extract the feature values ​​and match them with the preset dust feature library, output the dust type label and wear coefficient, and input the air permeability time series, ambient temperature and humidity, dust load and engine operating conditions according to the dust holding capacity, corrected pressure difference data, ambient temperature and humidity and engine operating conditions to obtain the remaining life percentage of the filter element, with a prediction error of <±8%;

[0060] Step 3: Based on the output remaining lifespan percentage and air permeability threshold, trigger a level 3 warning:

[0061] When the air permeability is ≤50%, a pop-up notification will be executed; when the air permeability is ≤40%, both pop-up and SMS notifications will be executed; when the air permeability is ≤20%, a screen pop-up reminder will be executed and the fuel consumption increment will be calculated.

[0062] Step 4: When the remaining lifespan of the output is less than 5%, send the filter element positioning data to the cloud. The cloud generates a reverse logistics work order. The company arranges after-sales personnel to visit the driver and recommend that the driver perform filter element maintenance. For filter elements with a residual value assessment of more than 30%, initiate the return-to-factory remanufacturing process.

[0063] In this embodiment of the invention, the filter element status parameters are accurately acquired through multi-sensor collaborative acquisition and data correction. Combined with dust type identification and LSTM life prediction model, the error of the remaining life prediction of the filter element is controlled within ±8%. Then, the intelligent triggering of maintenance is realized through a three-level early warning mechanism. Finally, when the filter element is scrapped, the reverse logistics and remanufacturing process is initiated, which effectively improves the accuracy and reliability of filter element status monitoring, reduces manual intervention, and reduces the risk of engine wear and resource waste.

[0064] In a preferred embodiment of the present invention, step 1 above may include:

[0065] Step 1.1: Using a MEMS laser scattering sensor, synchronously collect PM2.5 concentration, PM10 concentration, and dust holding capacity data at a fixed period of 10 seconds to obtain the collected data;

[0066] Step 1.2: Based on the collected data, outliers caused by instantaneous environmental interference are removed, the dust holding capacity is converted into dust load, and the verified PM2.5 or PM10 concentration and dust load are encrypted and uploaded to the cloud through a low-power communication module.

[0067] In this embodiment of the invention, a MEMS laser scattering sensor is used to synchronously collect PM2.5 or PM10 concentration and dust holding capacity data at a fixed 10-second cycle. Combined with the elimination of instantaneous environmental interference anomalies and the conversion of dust holding capacity to dust load, the verified data is transmitted through a low-power communication module. This enables high-frequency and accurate acquisition of filter dust concentration and load, effectively avoiding data distortion caused by environmental interference. It provides a reliable data foundation for subsequent dust type identification, lifespan prediction, and early warning. At the same time, the low-power design ensures the chip's battery life and improves the system's practicality and stability.

[0068] In this embodiment of the invention, the specific steps include:

[0069] Step 1.1: Using a MEMS laser scattering sensor, PM2.5 concentration, PM10 concentration, and dust holding capacity data are collected synchronously in a fixed cycle of 10 seconds to obtain the collected data.

[0070] Step 1.2: Perform sliding window (5 periods) statistics on the buffer data, calculate the mean (μ) and standard deviation (σ) of PM2.5 / PM10 concentrations, and remove outliers that meet the following conditions:

[0071] Concentration value > μ + 3σ or < μ - 3σ;

[0072] Dust holding capacity fluctuation > 50mg / cycle;

[0073] Dust holding capacity to dust load conversion.

[0074] The dust holding capacity data after cleaning is converted into the dust load on the filter element surface (unit: g). The calculation formula is as follows: ,in, Dust holding capacity (mg / m³) 3 (), take the sum of PM2.5 and PM10 concentrations, It is the sampling flow rate. It is the cumulative sampling time (h). It is the effective filtration area of ​​the filter element.

[0075] The PM2.5 / PM10 concentration and dust load are checked using CRC-16 to generate a check code. The data is then encrypted using the AES-128 algorithm, with the key dynamically distributed from the cloud.

[0076] In a preferred embodiment of the present invention, step 1 above may include:

[0077] Step 1.3: The temperature and relative humidity of the environment where the filter element is located are obtained in real time by the temperature and humidity sensor integrated on the chip. The original output value of the differential pressure sensor is input into the linear compensation formula, correlated with the dust load data, and high clogging risk areas are marked.

[0078] Step 1.4: Based on high congestion risk areas, when a sudden change in carbon dioxide concentration of a single vehicle is detected to be >15%, retrieve carbon dioxide data of other vehicles within 10 kilometers of the same area. If the proportion of vehicles with sudden changes in the area is >30%, it is determined to be an environmental factor. If only a single vehicle has a sudden change, correlate it with the engine operating data of that vehicle to determine a fuel injection failure.

[0079] In this embodiment of the invention, the ambient temperature and humidity are acquired in real time by a temperature and humidity sensor integrated on the chip. The raw data from the differential pressure sensor is linearly compensated and associated with the dust load to mark high-clogging risk areas. This effectively eliminates the interference of temperature and humidity on differential pressure monitoring and improves the accuracy of filter clogging status judgment. At the same time, through the regional analysis mechanism of carbon dioxide concentration mutation, environmental factors and engine fuel injection failures can be accurately distinguished, avoiding misjudgment and quickly locating the root cause of the problem. This provides a more reliable decision-making basis for filter status assessment and engine fault early warning, thereby reducing the risk of maintenance misoperation and engine wear.

[0080] In this embodiment of the invention, the specific steps include:

[0081] Step 1.3: Using the temperature and humidity sensor integrated on the chip, the temperature and relative humidity values ​​of the environment where the filter element is located are acquired in real time. The raw output value of the differential pressure sensor is then input into the linear compensation formula. ,in, =0.002 ℃, =0.001 %RH, Temperature (°C) Relative humidity (%) The original output value of the differential pressure sensor is used, along with the dust load data, to mark high-clogging risk areas. Temperature is normalized with 25℃ as the baseline, assuming a temperature fluctuation range of 0 to 50℃, and the temperature deviation is normalized to the range of [-1, 1]. Humidity is normalized with 50%RH as the baseline, assuming a humidity fluctuation range of 0 to 100%RH, and the humidity deviation is normalized to the range of [-1, 1].

[0082] Step 1.4: Based on the marked high-congestion-risk areas, when a sudden change in carbon dioxide concentration of a single vehicle exceeds 15%, carbon dioxide data of other vehicles within 10 kilometers of the same area are retrieved. If the proportion of vehicles with sudden changes in this area exceeds 30%, it is determined to be caused by environmental factors; if only a single vehicle experiences a sudden change, the engine operating data of that vehicle is correlated to determine whether it is a fuel injection failure.

[0083] In a preferred embodiment of the present invention, step 2 above may include:

[0084] Step 2.1: Based on the corrected differential pressure data and dust load, extract the feature values ​​of the laser scattering intensity matrix, including calculating the average, variance, and peak value of the scattering intensity as feature vectors; perform similarity matching between the feature vectors and the preset dust feature library, and output dust type labels; simultaneously, fuse differential pressure-dust coupled heat map data to mark the distribution areas of high-wear dust; when more than 50 vehicles in the same geographical area identify a new dust type that is not preset, automatically expand the categories of the dust feature library and send it to the edge computing unit for updating;

[0085] Step 2.2: Based on the dust type label, connect to the vehicle manufacturer's ECU historical data platform to retrieve the engine cylinder pressure fluctuation value during the filter's service life; calculate the wear coefficient based on the cylinder pressure fluctuation value, specifically by statistically analyzing the standard deviation of the cylinder pressure fluctuation to quantify the degree of wear, and output the wear coefficient in association with the dust type label.

[0086] In this embodiment of the invention, feature values ​​of the laser scattering intensity matrix are extracted based on the corrected differential pressure data and dust load. These feature values ​​are then matched with a preset dust feature library to output type labels. A differential pressure-dust coupled thermal map is fused to annotate high-wear areas. When more than 50 vehicles in the same geographical area identify a new dust type, the feature library is automatically expanded. This allows for accurate identification of dust types and dynamic adaptation to complex environments, avoiding excessive filter replacement or engine damage due to misjudgment of dust types. Simultaneously, based on the dust type labels, the system connects to the historical data platform of the vehicle manufacturer's ECU. By statistically analyzing the standard deviation of engine cylinder pressure fluctuations and correlating it with dust types, the wear degree can be quantified. This allows for accurate assessment of the wear impact of dust on the engine, providing more reliable wear data support for predicting the remaining life of the filter element. The filter element condition judgment error is reduced from ±25% in traditional solutions to ±8%, reducing filter element waste by more than 100,000 units per year and improving the accuracy and environmental adaptability of filter element condition monitoring.

[0087] In this embodiment of the invention, the specific steps include:

[0088] Step 2.1, extract the eigenvalues ​​of the laser scattering intensity matrix:

[0089] A correlation analysis was performed on the corrected differential pressure data and dust load to determine the effective data range of the laser scattering intensity matrix. The average value and variance of the scattering intensity in the matrix were calculated to characterize the dispersion and stability of the scattering intensity. The peak values ​​of the scattering intensity were identified and extracted as key features reflecting sudden high scattering events. The average value, variance, and peak values ​​were then combined into a feature vector. ,in, It is the average value of the scattering intensity. It is the variance of the scattering intensity. It is the peak value of laser scattering intensity, which serves as the basic data for dust type identification.

[0090] Dust type label matching and output:

[0091] The generated feature vector is compared with the feature vectors of various dust types in the preset dust feature library. The similarity is calculated and matched in descending order of similarity. The dust type with the highest similarity is selected as the matching result and the matching dust type label is output.

[0092] High-wear dust distribution area marking:

[0093] By integrating differential pressure data and dust load data, a differential pressure-dust coupled heat map is generated, with the horizontal axis representing differential pressure (kPa) and the vertical axis representing dust load (g). In the heat map, based on a preset wear threshold, areas with high dust concentration and significant differential pressure changes are identified, and these areas are marked to determine them as high-wear dust distribution areas.

[0094] Dynamic expansion of the dust feature library:

[0095] The system provides real-time statistics on the types of dust detected by vehicles within the same geographical area. When more than 50 vehicles within the same geographical area detect a new dust type not in the preset dust feature library, the feature library expansion mechanism is triggered. The newly detected dust feature vectors are added to the dust feature library to form a new category. The updated dust feature library is then distributed to the edge computing unit to achieve real-time updates of the local model.

[0096] Step 2.2: Based on the output dust type label, determine the filter element's usage cycle to be analyzed. Connect to the vehicle manufacturer's ECU historical data platform and retrieve engine cylinder pressure fluctuation data during the filter element's usage according to the determined cycle. Perform preliminary cleaning on the retrieved cylinder pressure fluctuation data to remove obviously abnormal values ​​and ensure data validity.

[0097] Wear coefficient calculation and quantification:

[0098] Statistical analysis was performed on the cylinder pressure fluctuation values ​​of the cleaned engine, and its standard deviation was calculated. This standard deviation was used as an indicator to quantify the degree of engine wear and was defined as the wear coefficient. The larger the standard deviation, the more severe the cylinder pressure fluctuation and the higher the degree of engine wear.

[0099] Output of wear coefficient and dust type label association:

[0100] The calculated wear coefficient is associated with the corresponding dust type label to form an associated data record containing the dust type label and the wear coefficient.

[0101] In a preferred embodiment of the present invention, step 2 above may include:

[0102] Step 2.3: Match the preset air permeability decay baseline curve according to the dust type label; adjust the slope of the decay curve by superimposing the wear coefficient; fuse the corrected differential pressure data and dust load to calculate the real-time air permeability decay value to obtain the initial value of remaining life; perform weighted fusion calculation with the regional working condition feature library and historical filter scrapping data, where the regional working condition feature weight accounts for 60% and the historical data weight accounts for 40%, and output the optimized value of remaining life percentage with a prediction error of <±8%;

[0103] Step 2.4: When the filter element is actually scrapped, dynamic compensation is performed based on the optimized remaining lifespan percentage output in Step 5.1.

[0104] The ratio of the predicted lifespan in days to the actual number of days used is calculated as an error compensation factor.

[0105] If the error compensation factor is greater than 1.1, add the attenuation rate compensation value for high dust conditions to the air permeability attenuation baseline curve.

[0106] If the error compensation factor is less than 0.9, a dust environment correction coefficient of 0.15 is added to the dust load.

[0107] In this embodiment of the invention, by matching a preset air permeability decay benchmark curve according to the dust type label and superimposing a wear coefficient correction slope, the real-time air permeability decay value is calculated by fusing the corrected differential pressure data and dust load. Then, the initial remaining lifespan value is weighted and fused with the regional operating condition feature library (weight 60%) and historical filter cartridge scrapping data (weight 40%) to accurately output the optimized remaining lifespan percentage value (prediction error < ±8%), achieving dynamic and accurate prediction of the filter cartridge's remaining lifespan. This significantly improves accuracy compared to the ±25% error of traditional technologies. When the filter cartridge is actually scrapped... The error compensation factor is calculated based on the ratio of predicted lifespan to actual lifespan. Dynamic compensation is applied to the air permeability decay baseline curve or dust load. This allows the prediction model to self-optimize according to actual usage scenarios, adapt to different regional working conditions and dust environments, avoid prediction deviations caused by a single baseline curve, further improve prediction accuracy, and reduce the risk of excessive filter replacement or engine damage due to inaccurate lifespan prediction. This reduces filter waste by more than 100,000 units per year and provides more reliable data support for filter residual value assessment and remanufacturing, promoting the intelligent and precise upgrading of filter management.

[0108] In this embodiment of the invention, the specific steps include:

[0109] Step 2.3, Match the air permeability decay baseline curve:

[0110] Based on the dust type labels output in step 2.2, access the built-in baseline curve library. The library structure is in key-value pair format: {dust type label: attenuation curve model}. The baseline curve library has 20 preset typical dust models, including:

[0111] Mining dust, road dust, metal particles, sea salt particles, cotton and linen fibers, mixed dust from construction sites, desert dust, industrial soot, pollen fibers, ceramic powder, etc.

[0112] Superimposed wear coefficient correction slope:

[0113] Use the wear coefficient calculated in step 2.2 to correct the slope of the reference curve. If the original slope of the reference curve is... The wear coefficient is Then the corrected slope .

[0114] Calculate the real-time air permeability decay value:

[0115] The permeability decay value is calculated by weighting the adjusted differential pressure data (ΔP) and dust load (D) after normalization. The normalization formula is: ,in, This is the corrected differential pressure data. =10kPa, =600kPa, Dust load, =0g, =300g.

[0116] Real-time air permeability decay value ,in, , For experience weights, usually =0.6, =0.4.

[0117] Get the initial value of remaining lifetime:

[0118] Based on the real-time air permeability decay value A, and referring to the air permeability-remaining lifetime mapping relationship corresponding to the baseline curve, the initial value of remaining lifetime L0 is obtained. For linear decay curves (such as silica dust), the mapping relationship can be expressed as: ,in, This is the real-time air permeability decay value. The air permeability reduction value when the filter element is scrapped (usually 80%). Initial filter decay value (usually 0%).

[0119] Weighted fusion generates optimized values:

[0120] Retrieve typical operating condition data (such as mountain roads, urban commuting, etc.) of the current region from the regional operating condition feature library; retrieve the average remaining life of the same type of dust from historical filter cartridge scrapping data.

[0121] Weighted formula for remaining lifetime optimization value: ,in, This is data from a regional operating condition characteristic database. Based on historical scrapping data, ensure that the prediction error is less than ±8%.

[0122] Step 2.4, calculate the error compensation factor:

[0123] When the filter element is actually scrapped, obtain the predicted lifespan days corresponding to the optimized remaining lifespan percentage output in step 2.3. and actual number of days used .

[0124] Error compensation factor .

[0125] Based on the factor-adjusted baseline curve:

[0126] If f > 1.1 (predicted lifespan is too long), in the permeability decay baseline curve, add a compensation value Δk to the decay rate for high dust conditions (dust load D > 200g). Compensation value .

[0127] If f < 0.9 (predicted lifespan is too short), add a dust load correction factor of 0.15 to the dust load D. The corrected load will be... Increase the weight of wear assessment in dusty environments.

[0128] Update the local prediction model:

[0129] The adjusted baseline curve or correction coefficient is synchronized to the edge computing unit to update the parameters of the local LSTM lifetime prediction model, thereby achieving model self-optimization.

[0130] In a preferred embodiment of the present invention, step 3 above may include:

[0131] Step 3.1: Read the breathability decay value corresponding to the remaining lifespan percentage optimization value in real time. When the breathability decay value reaches the initial threshold of 50%, activate the IoT communication module to send an APP pop-up and SMS notification to the user terminal.

[0132] Step 3.2: When the air permeability decay value continues to drop to 40%, the vehicle voice module is invoked to broadcast the warning content. At the same time, the current pressure difference increase ratio is calculated based on the corrected pressure difference data. If the pressure difference increase ratio is >30%, the warning level is simultaneously upgraded to level two.

[0133] Step 3.3: When the air permeability decay value reaches the critical threshold of 20%, a forced pop-up reminder is executed on the screen, and fuel consumption increment calculation is performed simultaneously.

[0134] Retrieve dust load data and output wear coefficient to calculate dust wear equivalent value;

[0135] Based on the data of the marked high-clogging risk areas, extract the actual value of the pressure difference before and after the current filter element;

[0136] By combining engine operating data, the actual pressure difference value is converted into the engine intake drag coefficient;

[0137] The absolute value of the fuel consumption increase per 100 kilometers is output by mapping the intake drag coefficient to the fuel consumption of the car manufacturer's ECU.

[0138] In this embodiment of the invention, by reading the air permeability decay value corresponding to the remaining lifespan percentage optimization value in real time, a three-level early warning mechanism is triggered in stages: when the air permeability decays to 50%, an APP pop-up window and SMS notification are sent to achieve early risk warning, so that the driver can pay attention to the status of the filter element in time; when it drops to 40%, the in-vehicle voice broadcast is invoked and the warning level is dynamically adjusted in combination with the pressure difference increase ratio to avoid misjudgment by a single indicator and improve the accuracy of the warning; when the critical threshold of 20% is reached, a forced pop-up window is executed on the screen and the fuel consumption increment is calculated (by the correlation calculation of the dust wear equivalent value, the actual pressure difference value and the engine intake resistance coefficient), so as to enhance the driver's initiative to replace the filter element in a way that quantifies the fuel consumption loss. This multi-level early warning system upgrades the vague early warning mode of traditional solutions to a data-driven, precise, and tiered alert system. This reduces maintenance response time from a delay of over 48 hours to real-time triggering. At the same time, it visualizes the economic impact of filter blockage through fuel consumption increments, encouraging drivers to proactively maintain their vehicles. This reduces engine repair costs caused by filter problems by more than 40% annually. Furthermore, the three-level early warning logic and pressure difference increase linkage mechanism improve the early warning accuracy to 92%, effectively avoiding excessive replacement and resource waste caused by delayed or misjudged early warnings.

[0139] In this embodiment of the invention, the specific steps include:

[0140] Step 3.1, Real-time data reading:

[0141] Read the air permeability decay value corresponding to the remaining lifetime percentage optimization value from the edge computing unit cache, with a read frequency of 1 time per minute.

[0142] Simultaneously retrieve the initial value of the current filter element's air permeability (the new filter element's air permeability is 100%) and the real-time decay curve.

[0143] Threshold determination and early warning activation:

[0144] When the air permeability decay value reaches or exceeds 50% for the first time, the first-level warning logic is triggered.

[0145] Activate the IoT communication module (Huawei HiSilicon Boudica200) to send warning commands to user terminals (mobile APP, vehicle central control screen) via 4G network.

[0146] Multi-terminal notification execution:

[0147] A red-bordered warning pop-up appeared in the FilterFu Smart Connect APP, displaying "Filter permeability decreased by 50%, it is recommended to pay attention to the status."

[0148] A text notification is sent to the driver's registered mobile phone number via SMS gateway. The notification includes the vehicle number, filter status, and maintenance suggestions.

[0149] Step 3.2, Secondary threshold determination:

[0150] Continuously monitor the air permeability decay value, and trigger a level 2 warning when it drops to 40%.

[0151] Synchronously retrieve the corrected differential pressure data from the past hour (sampling frequency 10 seconds per time).

[0152] Pressure differential increase calculation and early warning upgrade:

[0153] Calculate the current pressure difference Pressure difference compared to 1 hour ago Growth rate: Growth rate = , This represents the current pressure difference across the filter element. This is the pressure difference from 1 hour ago.

[0154] If the increase rate is greater than 30%, the warning level will be raised from Level 1 to Level 2; otherwise, Level 2 warning will remain in effect.

[0155] Voice broadcast and data synchronization:

[0156] The vehicle's voice module is invoked, and the announcement reads: "Filter permeability has decreased by 40%, and the current pressure difference has increased by XX%. Please arrange for an inspection as soon as possible."

[0157] A yellow warning icon is displayed on the vehicle's dashboard, and details of the level 2 warning are pushed to the app.

[0158] Step 3.3, mandatory early warning for critical thresholds:

[0159] When the air permeability decreases by 20%, a forced pop-up window is triggered, covering the current vehicle system interface and displaying a red warning icon and the text "Replace filter immediately".

[0160] Simultaneously lock some functions of the vehicle's central control screen until the driver confirms the warning information.

[0161] Calculation of dust abrasion equivalent value:

[0162] Retrieve current dust load ( (Unit: g) and the wear coefficient output from step 2.2 ( ).

[0163] Wear equivalent value = It is used to quantify the overall wear and tear of dust on the filter element and engine.

[0164] Intake drag coefficient conversion:

[0165] Extract the actual current pressure difference from the data labeled in high-congestion-risk areas. (Unit: kPa)

[0166] By combining real-time engine operating data (speed, load), the pressure difference is converted into the intake drag coefficient using a drag conversion formula. ): , Standard atmospheric pressure This is the engine drag coefficient constant.

[0167] Fuel consumption increment calculation and output:

[0168] Find the fuel consumption mapping table (k-ΔL / 100km) calibrated by the car manufacturer's ECU, input the intake drag coefficient k, and get the corresponding fuel consumption increment per 100 kilometers (ΔL).

[0169] The pop-up window displays: "Clogged filter element causes increased fuel consumption ΔLL / 100km. Immediate replacement can reduce losses."

[0170] In a preferred embodiment of the present invention, step 4 above may include:

[0171] Step 4.1: When the output remaining life percentage optimization value is lower than 5% for the first time, retrieve the real-time location data of the vehicle's GPS module, combine it with the filter serial number to generate a unique identifier, and upload the identifier and location data to the cloud server through the IoT communication module.

[0172] Step 4.2: Based on the received filter cartridge serial number, the company arranges after-sales personnel to follow up with the driver, recommending filter cartridge maintenance, and retrieving the output wear coefficient and recorded dynamic compensation historical data to perform residual value assessment calculation.

[0173] Extract the total cumulative dust load of the filter element and divide it by the preset maximum dust holding capacity threshold of the filter material to obtain the physical loss rate. Weight the wear coefficient and the physical loss rate, with weights of 70% and 30% respectively, and output the residual value percentage. If the residual value percentage is greater than 30%, automatically generate a reverse logistics work order containing filter element positioning information and simultaneously perform a remanufacturing feasibility assessment.

[0174] The calculated dust wear equivalent value is retrieved. If the wear equivalent value is ≤0.3 and the error compensation factor is in the range of [0.95, 1.05], a filter element disassembly instruction is sent to the remanufacturing plant. If the wear equivalent value is >0.3, the filter element scrapping process is triggered and the user is notified.

[0175] In this embodiment of the invention, when the remaining lifespan percentage optimization value first falls below 5%, the vehicle's GPS location data is retrieved and combined with the filter element serial number to generate a unique identifier code, which is then uploaded to the cloud. This enables precise location and traceability of the filter element's scrap location, providing a data foundation for reverse logistics. The cloud retrieves the wear coefficient, dynamic compensation historical data, and cumulative dust load based on the filter element serial number. The residual value is accurately assessed through a weighted fusion of physical loss rate and wear coefficient (70%:30%). When the residual value percentage is greater than 30%, a reverse logistics work order containing location information is automatically generated. The feasibility of remanufacturing is determined by a dual assessment of the wear equivalent value (≤0.3) and the error compensation factor ([0.95,1.05]), thereby increasing the filter element residual value utilization rate from 5% in traditional technology to 35%. This process not only enables intelligent recycling management of scrapped filter cartridges, avoiding the waste of resources caused by the direct disposal of 30% of the remaining lifespan of filter cartridges in a certain mining area, but also promotes a carbon emission reduction of 1.2 kg CO2 e per filter cartridge through precise residual value assessment and remanufacturing determination, helping to achieve the "dual carbon" target. At the same time, it shortens the reverse logistics response time from the unsystematic management of traditional solutions to real-time triggering, forming a circular economy closed loop of "monitoring-scrap-recycling-remanufacturing", which greatly improves the economy and environmental protection of filter cartridge life cycle management.

[0176] In this embodiment of the invention, the specific steps include:

[0177] Step 4.1, Remaining lifetime threshold monitoring:

[0178] The remaining lifetime percentage optimization value output by the edge computing unit is polled in real time, with a monitoring frequency of once every 10 minutes. When this value first drops below 5%, the scrapping process is triggered.

[0179] GPS location data retrieval:

[0180] Access the vehicle's GPS module via the onboard OBD interface to obtain real-time longitude, latitude, and altitude data (accuracy ±10 meters), and record the data collection timestamp (accurate to the second).

[0181] Unique identifier generation:

[0182] Read the 16-digit serial number (e.g., ABC123456789DEF) in the filter's built-in NFC chip.

[0183] The serial number and GPS coordinates are concatenated in the format of "serial number_longitude_latitude_timestamp" to generate a unique identifier.

[0184] Data is uploaded in encrypted form.

[0185] The identification code and location data are packaged into JSON format using an IoT communication module (supporting 4G / 5G), the data packet is encrypted using the AES-128 algorithm, and then uploaded to the designated interface of the cloud server.

[0186] Step 4.2, Historical Data Retrieval:

[0187] Query the cloud database based on the filter cartridge serial number to obtain: the wear coefficient (w) output in step 2.2, and the dynamic compensation historical data (error compensation factor f sequence) recorded in step 2.4.

[0188] Filter element's total dust load over its entire lifespan ( (Unit: g)

[0189] Physical loss rate calculation:

[0190] The preset maximum dust holding capacity of the filter material is 300g, and the physical loss rate is = ,in, The cumulative dust load (g) is the total dust load. The preset maximum dust holding capacity threshold (g).

[0191] Residual value percentage weighted calculation:

[0192] Wear coefficient weighted at 70%, physical loss rate weighted at 30%.

[0193] Residual value percentage = , The wear coefficient is denoted as .

[0194] Reverse logistics work order generation:

[0195] If the residual value percentage is greater than 30%, a reverse logistics work order will be automatically generated, including:

[0196] Filter element positioning information;

[0197] Residual value assessment report;

[0198] The estimated recovery time (default 48 hours) is when the work order is pushed to the third-party logistics platform via API.

[0199] Remanufacturing feasibility assessment:

[0200] Retrieve the dust abrasion equivalent value calculated in step 3.3.

[0201] Simultaneously obtain the error compensation factor f from the last dynamic compensation:

[0202] If E≤0.3 and f∈[0.95,1.05], a dismantling instruction is sent to the remanufacturing plant. If E>0.3, the scrapping process is triggered, and a scrapping notification SMS is sent to the user.

[0203] like Figure 2 As shown, embodiments of the present invention also provide an air filter IoT smart chip monitoring system, comprising:

[0204] The acquisition module is used to collect the PM2.5 or PM10 concentration and dust holding capacity of the filter element at 10-second intervals using a laser particulate sensor, monitor the pressure difference across the filter element using a differential pressure sensor, and correct the pressure difference data using a temperature and humidity compensation algorithm. It also detects the intake air concentration using a non-dispersive infrared carbon dioxide sensor and identifies abnormal air-fuel ratios where the carbon dioxide concentration changes by more than 15%.

[0205] The extraction module is used to extract feature values ​​and match them with a preset dust feature library based on dust holding capacity and laser scattering intensity matrix feature values, and output dust type labels and wear coefficients. Based on dust holding capacity, corrected differential pressure data, ambient temperature and humidity, and engine operating conditions, the module inputs air permeability time series, ambient temperature and humidity, dust load, and engine operating conditions to obtain the remaining life percentage of the filter element, with a prediction error of <±8%.

[0206] The calculation module triggers a three-level warning based on the output percentage of remaining lifespan and the air permeability threshold: when the air permeability is ≤50%, a pop-up notification is executed; when the air permeability is ≤40%, a pop-up and SMS notification are executed; when the air permeability is ≤20%, a screen pop-up reminder is executed and the fuel consumption increment is calculated.

[0207] When the output remaining lifespan is less than 5%, the processing module sends the filter location data to the cloud. The company then arranges after-sales personnel to visit the driver and advise the driver to perform filter maintenance. For filters with a residual value of more than 30%, the remanufacturing process is initiated.

[0208] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for monitoring air filter IoT smart chips, characterized in that, The method comprises: Step 1: Collect PM2.5 or PM10 concentration and dust holding capacity of the filter element through a laser particulate sensor; monitor the pressure difference across the filter element through a differential pressure sensor; correct the pressure difference data using a temperature and humidity compensation algorithm; detect the intake air concentration through a non-dispersive infrared carbon dioxide sensor; and identify abnormal air-fuel ratios where carbon dioxide concentration changes by more than 15%. Step 2: Based on the dust holding capacity and the laser scattering intensity matrix feature values, extract the feature values ​​and match them with the preset dust feature library, output the dust type label and wear coefficient, and input the air permeability time series, ambient temperature and humidity, dust load and engine operating conditions according to the dust holding capacity, corrected pressure difference data, ambient temperature and humidity and engine operating conditions to obtain the remaining life percentage of the filter element, with a prediction error of <±8%; Based on dust holding capacity, corrected differential pressure data, ambient temperature and humidity, and engine operating conditions, input the air permeability time series, ambient temperature and humidity, dust load, and engine operating conditions to obtain the remaining percentage of filter life. The prediction error is <±8%, including: Step 2.3: Match the preset air permeability decay baseline curve according to the dust type label; adjust the slope of the decay curve by superimposing the wear coefficient; fuse the corrected differential pressure data and dust load to calculate the real-time air permeability decay value to obtain the initial value of remaining life; perform weighted fusion calculation with the regional working condition feature library and historical filter scrapping data, where the regional working condition feature weight accounts for 60% and the historical data weight accounts for 40%, and output the optimized value of remaining life percentage with a prediction error of <±8%; Step 2.4: When the filter element is actually scrapped, dynamic compensation is performed based on the optimized remaining lifespan percentage output in Step 5.

1. The ratio of the predicted lifespan in days to the actual number of days used is calculated as an error compensation factor. If the error compensation factor is greater than 1.1, add the attenuation rate compensation value for high dust conditions to the air permeability attenuation baseline curve. If the error compensation factor is less than 0.9, a dust environment correction factor of 0.15 is added to the dust load. Step 3: Based on the output remaining lifespan percentage and air permeability threshold, trigger a level 3 warning: A pop-up notification will be executed when the air permeability is ≤50%. When the air permeability is ≤40%, a pop-up window and SMS notification will be executed. When the air permeability is ≤20%, a pop-up reminder will be displayed on the screen and the fuel consumption increment will be calculated. Step 4: When the remaining lifespan of the output is less than 5%, send the filter element positioning data to the cloud. The cloud generates a reverse logistics work order. The company arranges after-sales personnel to visit the driver and recommend that the driver perform filter element maintenance. For filter elements with a residual value assessment of more than 30%, initiate the return-to-factory remanufacturing process.

2. The method for monitoring an air filter using an IoT smart chip according to claim 1, characterized in that, The PM2.5 or PM10 concentration and dust holding capacity of the filter element are collected using a laser particulate sensor, including: Step 1.1: Using a MEMS laser scattering sensor, synchronously collect PM2.5 concentration, PM10 concentration, and dust holding capacity data at a fixed period of 10 seconds to obtain the collected data; Step 1.2: Based on the collected data, outliers caused by instantaneous environmental interference are removed, the dust holding capacity is converted into dust load, and the verified PM2.5 or PM10 concentration and dust load are encrypted and uploaded to the cloud through a low-power communication module.

3. The method for monitoring an air filter IoT smart chip according to claim 2, characterized in that, The pressure difference across the filter element is monitored by a differential pressure sensor, and the pressure difference data is corrected using a temperature and humidity compensation algorithm. Intake air concentration is detected by a non-dispersive infrared carbon dioxide sensor to identify air-fuel ratio anomalies where carbon dioxide concentration changes by more than 15%, including: Step 1.3: The temperature and relative humidity of the environment where the filter element is located are obtained in real time by the temperature and humidity sensor integrated on the chip. The original output value of the differential pressure sensor is input into the linear compensation formula, correlated with the dust load data, and high clogging risk areas are marked. Step 1.4: Based on high congestion risk areas, when a sudden change in CO2 concentration of a single vehicle is detected to be >15%, retrieve CO2 data of other vehicles within 10 kilometers of the same area. If the proportion of vehicles with sudden changes in the area is >30%, it is determined to be an environmental factor. If only a single vehicle has a sudden change, correlate it with the engine operating data of that vehicle to determine a fuel injection failure.

4. The method for monitoring an air filter IoT smart chip according to claim 3, characterized in that, Based on the dust holding capacity and laser scattering intensity matrix feature values, feature values ​​are extracted and matched against a pre-defined dust feature library, outputting dust type labels and wear coefficients, including: Step 2.1: Based on the corrected differential pressure data and dust load, extract the feature values ​​of the laser scattering intensity matrix, including calculating the average, variance, and peak value of the scattering intensity as feature vectors; perform similarity matching between the feature vectors and the preset dust feature library, and output dust type labels; simultaneously, fuse differential pressure-dust coupled heat map data to mark the distribution areas of high-wear dust; when more than 50 vehicles in the same geographical area identify a new dust type that is not preset, automatically expand the categories of the dust feature library and send it to the edge computing unit for updating; Step 2.2: Based on the dust type label, connect to the vehicle manufacturer's ECU historical data platform to retrieve the engine cylinder pressure fluctuation value during the filter's service life; calculate the wear coefficient based on the cylinder pressure fluctuation value, specifically by statistically analyzing the standard deviation of the cylinder pressure fluctuation to quantify the degree of wear, and output the wear coefficient in association with the dust type label.

5. The method for monitoring an air filter IoT smart chip according to claim 4, characterized in that, Based on the output remaining lifespan percentage and breathability threshold, a three-level warning is triggered: when breathability is ≤50%, a pop-up notification is executed; when breathability is ≤40%, both a pop-up notification and an SMS notification are executed. When the air permeability is ≤20%, a pop-up reminder will be displayed on the screen, and the increase in fuel consumption will be calculated, including: Step 3.1: Read the breathability decay value corresponding to the remaining lifespan percentage optimization value in real time. When the breathability decay value reaches the initial threshold of 50%, activate the IoT communication module to send an APP pop-up and SMS notification to the user terminal. Step 3.2: When the air permeability decay value continues to drop to 40%, the vehicle voice module is invoked to broadcast the warning content. At the same time, the current pressure difference increase ratio is calculated based on the corrected pressure difference data. If the pressure difference increase ratio is >30%, the warning level is simultaneously upgraded to level two. Step 3.3: When the air permeability decay value reaches the critical threshold of 20%, a forced pop-up reminder is executed on the screen, and fuel consumption increment calculation is performed simultaneously. Retrieve dust load data and output wear coefficient to calculate dust wear equivalent value; Based on the data of the marked high-clogging risk areas, extract the actual value of the pressure difference before and after the current filter element; By combining engine operating data, the actual pressure difference value is converted into the engine intake drag coefficient; The absolute value of the fuel consumption increase per 100 kilometers is output by mapping the intake drag coefficient to the fuel consumption of the car manufacturer's ECU.

6. The method for monitoring an air filter IoT smart chip according to claim 5, characterized in that, When the remaining lifespan of the output filter is less than 5%, the filter location data is sent to the cloud. The cloud generates a reverse logistics work order, and the company arranges after-sales personnel to follow up with the driver, advising the driver to perform filter maintenance. For filter elements with a residual value assessment greater than 30%, the return-to-factory remanufacturing process is initiated, including: Step 4.1: When the output remaining life percentage optimization value is lower than 5% for the first time, retrieve the real-time location data of the vehicle's GPS module, combine it with the filter serial number to generate a unique identifier, and upload the identifier and location data to the cloud server through the IoT communication module. Step 4.2: Based on the received filter cartridge serial number, the company arranges after-sales personnel to follow up with the driver, recommending filter cartridge maintenance, and retrieving the output wear coefficient and recorded dynamic compensation historical data to perform residual value assessment calculation. Extract the total cumulative dust load of the filter element and divide it by the preset maximum dust holding capacity threshold of the filter material to obtain the physical loss rate. Weight the wear coefficient and the physical loss rate, with weights of 70% and 30% respectively, and output the residual value percentage. If the residual value percentage is greater than 30%, automatically generate a reverse logistics work order containing filter element positioning information and simultaneously perform a remanufacturing feasibility assessment. The calculated dust wear equivalent value is retrieved. If the wear equivalent value is ≤0.3 and the error compensation factor is in the range of [0.95, 1.05], a filter element disassembly instruction is sent to the remanufacturing plant. If the wear equivalent value is >0.3, the filter element scrapping process is triggered and the user is notified.

7. An air filter IoT smart chip monitoring system, wherein the system implements the method as described in any one of claims 1 to 6, characterized in that, include: The acquisition module is used to collect the PM2.5 or PM10 concentration and dust holding capacity of the filter element at 10-second intervals using a laser particulate sensor, monitor the pressure difference across the filter element using a differential pressure sensor, and correct the pressure difference data using a temperature and humidity compensation algorithm. It also detects the intake air concentration using a non-dispersive infrared carbon dioxide sensor and identifies abnormal air-fuel ratios where the carbon dioxide concentration changes by more than 15%. The extraction module is used to extract feature values ​​and match them with a preset dust feature library based on dust holding capacity and laser scattering intensity matrix feature values, and output dust type labels and wear coefficients. Based on dust holding capacity, corrected differential pressure data, ambient temperature and humidity, and engine operating conditions, the module inputs air permeability time series, ambient temperature and humidity, dust load, and engine operating conditions to obtain the remaining life percentage of the filter element, with a prediction error of <±8%. The calculation module is used to trigger a three-level warning based on the output remaining lifespan percentage and air permeability threshold: when the air permeability is ≤50%, a pop-up notification is executed; when the air permeability is ≤40%, a pop-up and SMS notification are executed; when the air permeability is ≤20%, a screen pop-up reminder is executed and the fuel consumption increment is calculated. The processing module sends filter location data to the cloud when the remaining lifespan of the output is less than 5%. The cloud generates a reverse logistics work order, and the company arranges after-sales personnel to visit the driver and recommend that the driver perform filter maintenance. For filters with a residual value assessment of more than 30%, the return-to-factory remanufacturing process is initiated.

8. A computing device, characterized in that, include: one or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 6.

Images