Air conditioner filter element use cycle prediction method and device

Abstract

The application provides an air conditioner filter core use cycle prediction method and device, and relates to the technical field of vehicle-mounted air conditioners. The method comprises the following steps: obtaining signal data, weather data and filter core after-sales maintenance data of a target vehicle within a preset time period; inputting the signal data, weather data and filter core after-sales maintenance data into an air conditioner filter core use cycle prediction model, and using the air conditioner filter core use cycle prediction model to predict and output a filter core health degree of a current filter core at a current time; wherein the air conditioner filter core use cycle prediction model is obtained by training an initial regression model with filter core health degrees of replaced air conditioner filter cores in a training data set as labels; the filter core health degrees of the replaced air conditioner filter cores are obtained by calculating a performance data matrix composed of the basis weight, air resistance and purification efficiency of the replaced air conditioner filter cores; and the filter core health degree of the current filter core at the current time is converted into a predicted use cycle of the current filter core.

Description

Technical Field

[0001] This application relates to the field of vehicle air conditioning technology, and more specifically, to a method and device for predicting the service life of an air conditioning filter. Background Technology

[0002] Vehicle air conditioning systems primarily use filters to purify dust, exhaust fumes, and other pollutants to ensure air quality inside the vehicle. As the air conditioning system is used over time, the filtering performance of the air conditioning filter will gradually decrease, so it is necessary to replace the air conditioning filter in a timely manner.

[0003] Currently, methods for predicting the lifespan of vehicle cabin air filters include: collecting measured airflow through the cabin air filter at the vehicle's current fan speed setting, and comparing the measured airflow with the calibrated airflow to determine if the cabin air filter needs replacement. However, due to the influence of intake airflow, the accuracy of the measured airflow is low, resulting in low accuracy in predicting the cabin air filter's lifespan. Alternatively, methods can determine the need for cabin air filter replacement based on the duration of air conditioning use. However, since each vehicle's driving environment is different, the filtration performance of cabin air filters varies significantly. Furthermore, existing technologies assume that the residual value of replaced cabin air filters is zero, ignoring the actual residual value of cabin air filters. Therefore, the accuracy of predicting the cabin air filter's lifespan based on existing technologies is also low.

[0004] Therefore, improving the accuracy of predicting the lifespan of vehicle air conditioning filters is an urgent problem to be solved. Summary of the Invention

[0005] To improve the accuracy of predicting the service life of vehicle air conditioning filters, this application provides a method and apparatus for predicting the service life of air conditioning filters.

[0006] In a first aspect, embodiments of this application provide a method for predicting the service life of an air conditioning filter, including: The system acquires signal data of the target vehicle within a preset time period, weather data of the target vehicle during the preset time period, and after-sales maintenance data of the target vehicle's filter element. The preset time period is the time period from the start time of the target vehicle's current filter element to the current time. The signal data, weather data, and filter after-sales maintenance data are input into the air conditioning filter usage cycle prediction model. The air conditioning filter usage cycle prediction model is used to predict and output the current filter health status at the current time. The air conditioning filter usage cycle prediction model is obtained by training an initial regression model using the filter health status of replaced air conditioning filters in the training dataset as labels. The filter health status of replaced air conditioning filters is calculated by the performance data matrix of replaced air conditioning filters, which is used to indicate the filtration performance of replaced air conditioning filters. The performance data matrix of replaced air conditioning filters is a matrix composed of the weight, air resistance, and purification efficiency of replaced air conditioning filters. The current filter element's health status at the current moment is converted into the predicted usage cycle of the current filter element.

[0007] As an optional implementation of this application, the step of converting the current filter element's health status at the current moment into the predicted usage cycle of the current filter element includes: The filter health status is mapped to the initial predicted usage period of the current filter at the current moment by linear interpolation, and the predicted usage periods corresponding to multiple dates before the current date are obtained respectively. The predicted usage period of the air conditioning filter at the current moment is calculated by taking the average of the initial predicted usage period corresponding to the current moment and the predicted usage periods corresponding to multiple dates prior to the current moment.

[0008] As an optional implementation of this application, after converting the current filter's health status at the current moment into the predicted usage cycle of the current filter, the method further includes: The remaining usage days of the current filter element are calculated based on its predicted usage cycle and the number of days it has been used at the current moment, and then the remaining usage days are sent to the vehicle for display.

[0009] As an optional implementation of this application, after sending the remaining usage days to the vehicle terminal for display, the method further includes: Calculate the time interval between the current moment and the remaining number of days of use; If the time interval is less than or equal to a preset time interval, a reminder message will be displayed on the vehicle side to remind the user to replace the air conditioning filter.

[0010] As an optional implementation of this application, before acquiring the signal data of the target vehicle within a preset time period, the weather data of the target vehicle during the preset time period, and the filter after-sales maintenance data of the target vehicle, the method further includes: training the initial regression model. Training the initial regression model includes: For multiple vehicles, sample data for each of the multiple vehicles is obtained, wherein the sample data for each vehicle includes: signal data of the vehicle during the period when the air conditioning filter has been replaced, weather data of the vehicle during the period when the air conditioning filter has been replaced, after-sales maintenance data of the vehicle's filter, and the filter health of the vehicle's replaced air conditioning filter. The signal data of the multiple vehicles during the period when the air conditioning filter was replaced, the weather data of the multiple vehicles during the period when the air conditioning filter was replaced, and the after-sales maintenance data of the multiple vehicles were used as feature sets, and the filter health of the replaced air conditioning filter of the multiple vehicles was used as a label to construct a training dataset. The feature set in the training dataset is input into the initial regression model. The parameters of the initial regression model are adjusted based on the predicted values ​​and labels output by the initial regression model until the training termination condition is met, thus obtaining the air conditioner filter usage cycle prediction model.

[0011] As an optional implementation of this application, obtaining sample data for each of the plurality of vehicles includes: Obtain the filter health status of the replaced air conditioning filter for each vehicle; Specifically, for each vehicle, the health status of the replaced air conditioning filter is obtained, including: Multiple replaced air conditioning filters of the vehicle are collected, and each replaced air conditioning filter is tested to obtain the weight, wind resistance and remaining purification efficiency of each replaced air conditioning filter. The difference between 1 and the remaining purification efficiency of each replaced air conditioning filter is taken as the purification efficiency of the corresponding replaced air conditioning filter. The performance data matrix is ​​constructed by using the weight, air resistance, and purification efficiency of each replaced air conditioning filter as row elements. The performance data matrix is ​​subjected to dimensionality reduction and weighting to obtain the residual filtration performance index of the replaced air conditioning filter of the vehicle. The difference between 1 and the residual filtration performance index of the replaced air conditioning filter of the vehicle is used as the filter health of the replaced air conditioning filter of the vehicle.

[0012] As an optional implementation of this application, the step of performing dimensionality reduction and weighting processing on the performance data matrix to obtain the residual filtration performance index of the replaced air conditioning filter of the vehicle includes: Normalize each element of the performance data matrix and calculate the covariance matrix of the normalized performance data matrix; Calculate the eigenvalues ​​of the covariance matrix and the eigenvectors corresponding to each eigenvalue; Arrange the eigenvectors corresponding to each eigenvalue from top to bottom according to the eigenvalues ​​in descending order to obtain the eigenma matrix corresponding to the performance data matrix; By using the eigenvalues ​​to weight each column of the matrix obtained by multiplying the feature matrix and the performance data matrix, the residual filtration performance index of the replaced air conditioning filter of the vehicle is obtained.

[0013] As an optional implementation of this application, the step of constructing a training dataset by using signal data of the multiple vehicles during the period of use of the replaced air conditioning filters, weather data of the multiple vehicles during the period of use of the replaced air conditioning filters, and after-sales maintenance data of the multiple vehicles' filters as a feature set, and using the filter health status of the replaced air conditioning filters of the multiple vehicles as a label, includes: The signal data of each vehicle during the period when the air conditioning filter has been replaced, the weather data of the vehicle during the period when the air conditioning filter has been replaced, and the after-sales maintenance data of the filter of the vehicle are ranked by importance to obtain the feature data ranking result. Data with importance lower than a preset importance in the sorted feature data are removed to obtain the filtered feature data; The selected feature data of the multiple vehicles are used as the feature set, and the health status of the replaced air conditioning filters of the multiple vehicles is used as the label to construct a training dataset.

[0014] Secondly, embodiments of this application provide an air conditioning filter life prediction device, comprising: The acquisition module is used to acquire signal data of the target vehicle within a preset time period, weather data of the target vehicle during the driving within the preset time period, and filter after-sales maintenance data of the target vehicle. The preset time period is the time period from the start time of the current filter of the target vehicle to the current time. The prediction module is used to input the signal data, the weather data, and the filter after-sales maintenance data into the air conditioning filter usage cycle prediction model, and use the air conditioning filter usage cycle prediction model to predict and output the filter health status of the current filter at the current time; wherein, the air conditioning filter usage cycle prediction model is obtained by training an initial regression model using the filter health status of replaced air conditioning filters in the training dataset as labels, and the filter health status of replaced air conditioning filters is calculated by the performance data matrix of replaced air conditioning filters, which is used to indicate the filtration performance of replaced air conditioning filters, and the performance data matrix of replaced air conditioning filters is a matrix composed of the weight, air resistance, and purification efficiency of replaced air conditioning filters; The conversion module is used to convert the current filter's health status at the current moment into the predicted usage cycle of the current filter.

[0015] As an optional implementation of this application, the conversion module is specifically used to map the filter health status to the initial predicted usage period of the current filter at the current moment through linear interpolation, and to obtain the predicted usage periods corresponding to multiple dates before the current moment. The predicted usage period of the air conditioning filter at the current moment is calculated by taking the average of the initial predicted usage period corresponding to the current moment and the predicted usage periods corresponding to multiple dates prior to the current moment.

[0016] As an optional implementation of this application, the device further includes: a display module, configured to, after converting the current filter's filter health status at the current moment into the predicted usage cycle of the current filter, calculate the remaining usage days of the current filter based on the predicted usage cycle of the current filter at the current moment and the number of days already used, and send the remaining usage days to the vehicle for display.

[0017] As an optional implementation of this application, the display module is used to calculate the time interval between the current time and the remaining usage days after sending the remaining usage days to the vehicle terminal for display; If the time interval is less than or equal to a preset time interval, a reminder message will be displayed on the vehicle side to remind the user to replace the air conditioning filter.

[0018] As an optional implementation of this application, the device further includes: a training module, used to train the initial regression model before acquiring the signal data of the target vehicle within a preset time period, the weather data of the target vehicle during the preset time period, and the filter after-sales maintenance data of the target vehicle. Training the initial regression model includes: For multiple vehicles, sample data for each of the multiple vehicles is obtained, wherein the sample data for each vehicle includes: signal data of the vehicle during the period when the air conditioning filter has been replaced, weather data of the vehicle during the period when the air conditioning filter has been replaced, after-sales maintenance data of the vehicle's filter, and the filter health of the vehicle's replaced air conditioning filter. The signal data of the multiple vehicles during the period when the air conditioning filter was replaced, the weather data of the multiple vehicles during the period when the air conditioning filter was replaced, and the after-sales maintenance data of the multiple vehicles were used as feature sets, and the filter health of the replaced air conditioning filter of the multiple vehicles was used as a label to construct a training dataset. The feature set in the training dataset is input into the initial regression model. The parameters of the initial regression model are adjusted based on the predicted values ​​and labels output by the initial regression model until the training termination condition is met, thus obtaining the air conditioner filter usage cycle prediction model.

[0019] As an optional implementation of this application, the acquisition module is specifically used to acquire the filter health of the replaced air conditioning filter of each vehicle. Specifically, for each vehicle, the health status of the replaced air conditioning filter is obtained, including: Multiple replaced air conditioning filters of the vehicle are collected, and each replaced air conditioning filter is tested to obtain the weight, wind resistance and remaining purification efficiency of each replaced air conditioning filter. The difference between 1 and the remaining purification efficiency of each replaced air conditioning filter is taken as the purification efficiency of the corresponding replaced air conditioning filter. The performance data matrix is ​​constructed by using the weight, air resistance, and purification efficiency of each replaced air conditioning filter as row elements. The performance data matrix is ​​subjected to dimensionality reduction and weighting to obtain the residual filtration performance index of the replaced air conditioning filter of the vehicle. The difference between 1 and the residual filtration performance index of the replaced air conditioning filter of the vehicle is used as the filter health of the replaced air conditioning filter of the vehicle.

[0020] As an optional implementation of this application, the acquisition module is specifically used to normalize each row of the performance data matrix and calculate the covariance matrix of the normalized performance data matrix. Calculate the eigenvalues ​​of the covariance matrix and the eigenvectors corresponding to each eigenvalue; Arrange the eigenvectors corresponding to each eigenvalue from top to bottom according to the eigenvalues ​​in descending order to obtain the eigenma matrix corresponding to the performance data matrix; By using the eigenvalues ​​to weight each column of the matrix obtained by multiplying the feature matrix and the performance data matrix, the residual filtration performance index of the replaced air conditioning filter of the vehicle is obtained.

[0021] As an optional implementation of this application, the training module is specifically used to sort the signal data of each vehicle during the period when the air conditioning filter has been replaced, the weather data of the vehicle during the period when the air conditioning filter has been replaced, and the after-sales maintenance data of the filter of the vehicle by importance, so as to obtain the feature data sorting result. Data with importance lower than a preset importance in the sorted feature data are removed to obtain the filtered feature data; The selected feature data of the multiple vehicles are used as the feature set, and the health status of the replaced air conditioning filters of the multiple vehicles is used as the label to construct a training dataset.

[0022] Thirdly, embodiments of this application provide an electronic device, including: a memory and a processor, wherein the memory is used to store a computer program, and the processor is used to execute the air conditioning filter usage cycle prediction method described in the first aspect or any optional embodiment of the first aspect when the computer program is invoked.

[0023] Fourthly, embodiments of this application provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the air conditioning filter life prediction method described in the first aspect or any optional implementation of the first aspect.

[0024] Fifthly, embodiments of this application provide a vehicle equipped with an air conditioning filter usage cycle prediction device as described in the third aspect, or an electronic device as described in the fifth aspect, or a storage medium as described in the sixth aspect.

[0025] The technical solution provided in this application has the following advantages compared with the prior art: This application provides a method and apparatus for predicting the usage cycle of an air conditioning filter. The method includes: acquiring signal data of a target vehicle within a preset time period, weather data of the target vehicle during the preset time period, and after-sales maintenance data of the target vehicle's filter, wherein the preset time period is the time between the start time of the current filter's use and the current time; inputting the signal data, the weather data, and the filter's after-sales maintenance data into an air conditioning filter usage cycle prediction model; and using the air conditioning filter usage cycle prediction model to predict and output the usage cycle of the current filter during the current time period. The air conditioning filter lifespan prediction model is obtained by training an initial regression model using the health status of replaced air conditioning filters in the training dataset as labels. The health status of replaced air conditioning filters is calculated from the performance data matrix of the replaced air conditioning filters, indicating their filtration performance. The performance data matrix of the replaced air conditioning filters is a matrix composed of the weight, air resistance, and purification efficiency of the replaced air conditioning filters. The current filter health status at the current moment is converted into the predicted lifespan of the current filter. On the one hand, since the data input to the air conditioning filter lifespan prediction model almost completely covers the influencing factors of air conditioning filter health, the prediction of the air conditioning filter lifespan no longer simply depends on mileage and time, greatly improving the prediction accuracy. On the other hand, since the health of the replaced air conditioning filter is used as a label when training the initial regression model, and the health of the replaced air conditioning filter is based on the weight, air resistance and purification efficiency of the replaced air conditioning filter, using the filter health as a label to train the initial regression model can improve the accuracy of the obtained air conditioning filter life prediction model, and further improve the accuracy of the current filter life prediction. Attached Figure Description The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0026] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a flowchart of an air conditioning filter life prediction method provided according to one or more embodiments of this application; Figure 2 This is a flowchart of an air conditioning filter life prediction method provided according to one or more embodiments of this application; Figure 3 A structural block diagram of an air conditioning filter life prediction device provided in one or more embodiments of this application; Figure 4 A structural block diagram of an air conditioning filter life prediction device provided in one or more embodiments of this application; Figure 5 This is an internal structural diagram of an electronic device provided according to one or more embodiments of this application. Detailed Implementation

[0028] To make the objectives, implementation methods and advantages of this application clearer, the exemplary implementation methods of this application will be clearly and completely described below with reference to the accompanying drawings of the exemplary embodiments of this application. Obviously, the described exemplary embodiments are only some embodiments of this application, and not all embodiments.

[0029] Based on the exemplary embodiments described in this application, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of the appended claims. Furthermore, although the disclosure in this application is presented by way of one or more exemplary examples, it should be understood that each aspect of these disclosures can constitute a complete implementation on its own. It should be noted that the brief descriptions of terminology in this application are merely for the convenience of understanding the embodiments described below, and are not intended to limit the implementation of this application. Unless otherwise stated, these terms should be understood in their ordinary and common meaning.

[0030] The air conditioning filter usage cycle prediction method provided in this application embodiment can be executed by the electronic device provided in this application embodiment, or it can be implemented by the air conditioning filter usage cycle prediction device provided in this application embodiment, or it can be implemented by one or more functional entities on the vehicle. This application embodiment does not make any specific limitations.

[0031] The following detailed description of the air conditioning filter life prediction method provided in this application is illustrated through several specific embodiments.

[0032] Figure 1 The flowchart of the air conditioning filter life prediction method provided in the embodiments of this application is shown below. Figure 1 As shown, the air conditioning filter usage cycle prediction method provided in this embodiment includes the following steps: S11. Obtain the signal data of the target vehicle within a preset time period, the weather data of the target vehicle during the preset time period, and the filter after-sales maintenance data of the target vehicle.

[0033] The preset time period is the period from the start time of the current filter of the target vehicle to the current time. The current filter is the air conditioning filter currently in use by the vehicle.

[0034] Signal data may include: air conditioning usage time, air conditioning circulation mode (including internal circulation mode and external circulation mode), fan speed, PM2.5 concentration inside the vehicle, vehicle speed, and cities traveled, etc.; weather data: temperature, air quality index (PM2.5, PM10, etc.), wind speed, humidity, rainfall, etc. of the target vehicle during the current filter usage period; filter after-sales maintenance data includes: the target vehicle's VIN, filter replacement time, vehicle mileage, etc.

[0035] The start time of current filter use can be understood as the installation time of current filter. For example, taking the installation time of current filter as the starting time, the following information is obtained for the target vehicle's air conditioning usage time, air conditioning circulation mode, fan speed, PM2.5 concentration in the vehicle, vehicle speed, and cities traveled during the time period up to the current time: temperature, air quality index (PM2.5, PM10, etc.), wind speed, humidity, rainfall, etc., during the time the target vehicle was traveling, as well as the target vehicle identification number (VIN), filter replacement time, vehicle mileage, etc.

[0036] S12. Input the signal data, the weather data, and the filter after-sales maintenance data into the air conditioning filter usage cycle prediction model, and use the air conditioning filter usage cycle prediction model to predict and output the filter health status of the current filter at the current time.

[0037] The air conditioning filter usage cycle prediction model is obtained by training an initial regression model using the filter health status of replaced air conditioning filters in the training dataset as labels. The filter health status of the replaced air conditioning filters is calculated by the performance data matrix of the replaced air conditioning filters, which is used to indicate the filtration performance of the replaced air conditioning filters. The performance data matrix of the replaced air conditioning filters is a matrix composed of the weight, air resistance, and purification efficiency of the replaced air conditioning filters.

[0038] For example, the calculation method for filter health may include the following process: For each vehicle, obtain the filter health of the replaced air conditioning filter of the vehicle, including: collecting multiple replaced air conditioning filters of the vehicle, testing each replaced air conditioning filter to obtain the weight, air resistance, and remaining purification efficiency of each replaced air conditioning filter, and taking the difference between 1 and the remaining purification efficiency of each replaced air conditioning filter as the purification efficiency of the corresponding replaced air conditioning filter; constructing the performance data matrix by using the weight, air resistance, and purification efficiency of each replaced air conditioning filter as row elements; and normalizing each row element of the performance data matrix. The system calculates the covariance matrix of the normalized performance data matrix; it calculates the eigenvalues ​​of the covariance matrix and the eigenvectors corresponding to each eigenvalue; it arranges the eigenvectors corresponding to each eigenvalue from top to bottom according to the eigenvalues ​​in descending order to obtain the feature matrix corresponding to the performance data matrix; it uses the eigenvalues ​​to weight each column of the matrix obtained by multiplying the feature matrix and the performance data matrix to obtain the residual filtration performance index of the replaced air conditioning filter of the vehicle; it calculates the difference between 1 and the residual filtration performance index of the replaced air conditioning filter of the vehicle as the filter health of the replaced air conditioning filter of the vehicle.

[0039] The health status of the replaced air conditioning filter is used as a label to train the initial regression model, resulting in an air conditioning filter usage cycle prediction model. Then, the target vehicle's air conditioning usage time, air conditioning circulation mode, fan speed, PM2.5 concentration inside the vehicle, vehicle speed, and the cities it traveled through are input into the air conditioning filter usage cycle prediction model. The target vehicle's temperature, air quality index (PM2.5, PM10, etc.), wind speed, humidity, rainfall, etc., during the same period are also input into the air conditioning filter usage cycle prediction model. Based on the air conditioning filter usage cycle prediction model, the current filter's health status at the current time is predicted and output.

[0040] Since the health status of replaced air conditioning filters was used as a label when training the initial regression model, and the health status of replaced air conditioning filters is based on the weight, air resistance, and purification efficiency of replaced air conditioning filters, using filter health status as a label to train the initial regression model can improve the accuracy of the obtained air conditioning filter life prediction model, making the predicted life more accurate.

[0041] S13. Convert the current filter element's filter element health status at the current moment into the predicted usage cycle of the current filter element.

[0042] For example, the health status of the filter element is mapped by linear interpolation to obtain the predicted usage cycle of the current filter element.

[0043] For example, the filter health can be mapped to the predicted usage cycle of the current filter at the current moment using the following formula (1): Formula (1); in, This represents the number of days the current filter cartridge has been used. This is the predicted lifespan of the current filter cartridge. This represents the initial health status of the current filter element, with a value of 1. The current filter element's health status at the current moment.

[0044] By input , ,as well as This allows us to obtain the predicted lifespan of the air conditioning filter at the current moment. To ensure that the model's prediction results meet the performance limits of professional evaluation, the prediction usage cycle can be understood as the prediction replacement cycle.

[0045] Among them, it is possible to Add restrictions, such as .

[0046] To ensure that the maintenance information received by users is relatively stable, the volatility of the daily forecast output of the air conditioning filter usage cycle prediction model can be considered. The MA moving average method can be used to average the time series of the result values ​​over a window of n and replace the original forecast values.

[0047] For example, the filter health is mapped to the initial predicted usage period of the current filter at the current moment by linear interpolation, and the predicted usage periods corresponding to multiple dates before the current moment are obtained respectively; the average of the initial predicted usage period corresponding to the current moment and the predicted usage periods corresponding to multiple dates before the current moment is calculated as the predicted usage period of the air conditioning filter at the current moment.

[0048] For example, the predicted service life of the air conditioning filter at the current moment can be calculated based on the following formula (2): Formula (2) in, The initial prediction period for the current moment. is the predicted lifespan of the air conditioning filter at the current moment, and n is the number of days in the moving average window.

[0049] Furthermore, based on the predicted usage cycle and the number of days already used of the current filter element at the current moment, the remaining usage days of the current filter element are calculated, and the remaining usage days are sent to the vehicle for display.

[0050] For example, by - This will give you the current number of days remaining in use.

[0051] Furthermore, the time interval between the current moment and the remaining number of days of use can be calculated; if the time interval is less than or equal to a preset time interval, a reminder message is displayed on the vehicle, which is used to remind the user to replace the air conditioning filter.

[0052] The notification message can be a text message, an application push notification, or the like.

[0053] In some embodiments, the vehicle's HU interface provides users with an option to "enable intelligent maintenance". If the user selects "yes", the user's driving location information is collected, and the weather data of the day is matched with the geographical location and driving time as features to participate in the model calculation. If the user selects "no", the user's driving location information is not collected, and only vehicle signal data and filter replacement information are used as features to participate in the model calculation.

[0054] This application provides a method for acquiring signal data of a target vehicle within a preset time period, weather data of the target vehicle during the preset time period, and after-sales maintenance data of the target vehicle's filter element. The preset time period is the period from the start time of the current filter element to the current time. The signal data, weather data, and after-sales maintenance data are input into an air conditioning filter element usage cycle prediction model. The model predicts and outputs the filter element health status of the current filter element at the current time. The air conditioning filter element usage cycle prediction model is obtained by training an initial regression model using the filter element health status of replaced air conditioning filters in the training dataset as labels. The filter element health status of replaced air conditioning filters is calculated by analyzing the performance data matrix of replaced air conditioning filters, which indicates the filtration performance of the replaced air conditioning filters. The performance data matrix of replaced air conditioning filters is a matrix composed of the weight, wind resistance, and purification efficiency of the replaced air conditioning filters. The filter element health status of the current filter element at the current time is converted into a predicted usage cycle of the current filter element. On the one hand, since the data input into the air conditioning filter lifespan prediction model almost completely covers the factors influencing the health of the air conditioning filter, the prediction of the air conditioning filter lifespan no longer simply relies on mileage and time, greatly improving the prediction accuracy. On the other hand, since the health of replaced air conditioning filters is used as a label when training the initial regression model, and the health of replaced air conditioning filters is based on their weight, air resistance, and purification efficiency, using filter health as a label to train the initial regression model can improve the accuracy of the resulting air conditioning filter lifespan prediction model, further improving the accuracy of the current filter's predicted lifespan.

[0055] Figure 2 A flowchart of an air conditioning filter life prediction method provided in another embodiment of this application is shown. Figure 1 Based on the illustrated embodiment, before step S11, a training process for the initial regression model is also included, exemplarily including the following steps S21 to S23.

[0056] S21. For multiple vehicles, obtain sample data for each of the multiple vehicles.

[0057] The sample data for each vehicle includes: signal data of the vehicle during the period when the air conditioning filter has been replaced, weather data of the vehicle during the period when the air conditioning filter has been replaced, after-sales maintenance data of the vehicle's filter, and the health status of the replaced air conditioning filter of the vehicle.

[0058] The health status of the replaced air conditioning filter is used to indicate the residual filtration performance index of the replaced air conditioning filter of the vehicle. The signal data and weather data can be continuously collected daily. After the air conditioning filter is replaced, the after-sales maintenance data and filter health status of the replaced air conditioning filter are collected at one time.

[0059] The replaced air conditioning filter refers to at least one air conditioning filter that has been used and replaced in the vehicle. The signal data of the vehicle during the period when the replaced air conditioning filter was in use may include: air conditioning usage duration, air conditioning circulation mode (including internal and external circulation modes), fan speed, PM2.5 concentration inside the vehicle, vehicle speed, and cities traveled, etc.; weather data of the vehicle during the period when the replaced air conditioning filter was in use: temperature, air quality index (PM2.5, PM10, etc.), wind speed, humidity, rainfall, etc.; and filter after-sales maintenance data of the vehicle including: vehicle identification number (VIN), filter replacement time, vehicle mileage, etc. Signal data and weather data can be continuously collected daily, while filter after-sales maintenance data and the health status of the replaced air conditioning filter can be collected simultaneously.

[0060] For each vehicle, obtaining the filter health of the replaced cabin air filters can include: collecting multiple replaced cabin air filters from the vehicle; testing each replaced cabin air filter to obtain its weight, air resistance, and remaining purification efficiency; using the difference between 1 and the remaining purification efficiency of each replaced cabin air filter as the purification efficiency of the corresponding replaced cabin air filter; constructing the performance data matrix using the weight, air resistance, and purification efficiency of each replaced cabin air filter as row elements; performing dimensionality reduction and weighting on the performance data matrix to obtain the residual filtration performance index of the replaced cabin air filters of the vehicle; and calculating the difference between 1 and the residual filtration performance index of the replaced cabin air filters of the vehicle as the filter health of the replaced cabin air filters of the vehicle.

[0061] The purification efficiency refers to the purification efficiency for PM2.5, and the air resistance can include air resistance at at least two different airflow rates, such as the air resistance corresponding to 200 cmh and 500 cmh. The remaining purification efficiency is the purification efficiency measured in the experiment. By subtracting the remaining purification efficiency of the replaced air conditioning filter from 1, the purification efficiency, which has the same meaning as the weight and air resistance of the replaced air conditioning filter, can be obtained.

[0062] The process of reducing the dimensionality and weighting the performance data matrix to obtain the residual filtration performance index of the replaced air conditioning filter of the vehicle may include: normalizing each row of the performance data matrix and calculating the covariance matrix of the normalized performance data matrix; calculating the eigenvalues ​​of the covariance matrix and the eigenvectors corresponding to each eigenvalue; arranging the eigenvectors corresponding to each eigenvalue from top to bottom according to the eigenvalues ​​in descending order to obtain the feature matrix corresponding to the performance data matrix; and weighting each column of the matrix obtained by multiplying the feature matrix and the performance data matrix using the eigenvalues ​​to obtain the residual filtration performance index of the replaced air conditioning filter of the vehicle.

[0063] For example, users are stratified and sampled according to region, mileage, and usage time, and brand new, unused filter cartridges are collected. The collected filter cartridges are then subjected to laboratory performance testing to obtain their corresponding weight, air resistance, and purification efficiency. The filter cartridge health is calculated as follows: The performance test data (weight, air resistance, and purification efficiency) are arranged into an n*m matrix X, where n is the number of data items and m is the number of filter cartridges; each row of X is normalized to its maximum and minimum values, transforming the performance test data into values ​​within the range [0,1].

[0064] Find the covariance matrix: .

[0065] Calculate the eigenvalues ​​of the covariance matrix ( ) and the corresponding feature vectors; arrange the feature vectors into matrix P from top to bottom according to the size of the corresponding feature values; Y=PX is the data after the feature vector transformation, and multiply each column of Y with the feature value as the weight to obtain W, which is the residual performance score (residual filtration performance index) after dimensionality reduction and weighting for each sample; perform feature transformation on the residual filtration performance index [0,1], and filter health S=1-W.

[0066] In this embodiment, the replaced air conditioning filter is tested to obtain its weight, air resistance, and purification efficiency. PCA modeling is then used to extract information corresponding to each of these parameters. This information is used as feature weights, and the multi-dimensional features are normalized based on these weights, for example, by using a weighted average to obtain a one-dimensional residual filtration performance index. The residual filtration performance index is a decimal between 0 and 1. If the residual filtration performance index is 0.4, then the filter health of the replaced air conditioning filter is 0.6.

[0067] S22. Using the signal data of the multiple vehicles during the period when the air conditioning filter has been replaced, the weather data of the multiple vehicles during the period when the air conditioning filter has been replaced, and the after-sales maintenance data of the multiple vehicles as feature sets, and using the filter health of the replaced air conditioning filter of the multiple vehicles as labels, a training dataset is constructed.

[0068] For example, when constructing a training dataset based on the aforementioned signal data, weather data, filter after-sales maintenance data, and filter health, some embodiments require preprocessing such as data cleaning. This includes cleaning data with issues such as missing values, outliers, erroneous values, data formats, and sampling rates, as well as converting data formats, such as converting continuous variables, discrete variables, and time series.

[0069] The preprocessed feature data is ranked by importance to remove low-importance data. For example, the signal data, weather data, and filter after-sales maintenance data of each vehicle during the period when the air conditioning filter was replaced are ranked by importance to obtain a feature data ranking result. Data with importance lower than a preset importance in the feature data ranking result is removed to obtain filtered feature data. The filtered feature data of the multiple vehicles is used as a feature set, and the filter health of the replaced air conditioning filters of the multiple vehicles is used as a label to construct a training dataset.

[0070] For example, the feature data shown in Table 1 below are sorted in descending order of importance. Assuming that the preset importance is the importance corresponding to the ranking of 30, the feature data in the ranking results are retained in the top 30 and the data ranked from 31 to 41 are removed.

[0071] Table 1

[0072] The features collected include: air conditioning usage time, air conditioning circulation mode, fan speed, PM2.5 concentration inside the vehicle, vehicle speed, and cities traveled through. The temperature, air quality index, wind speed, humidity, and rainfall during the use of the replaced air conditioning filter, vehicle identification number (VIN), filter replacement time, and vehicle mileage, etc.

[0073] S23. Input the feature set in the training dataset into the initial regression model, and adjust the parameters of the initial regression model based on the predicted value and label output by the initial regression model until the training termination condition is met, thereby obtaining the air conditioner filter usage cycle prediction model.

[0074] The parameters of the initial regression model can be adjusted by calculating the mean squared error, mean absolute error, and coefficient of determination between the predicted value output by the initial regression model and the label.

[0075] For example, the parameters of the initial regression model are adjusted so that the mean squared error (MSE) determined based on the predicted value and the label is less than the preset MSE, the mean absolute error (MAE) is less than the preset MAE, and the coefficient of determination R... 2 Less than the preset R 2 If the training termination condition is met, the model training is stopped, and the air conditioner filter usage cycle prediction model is obtained.

[0076] In this embodiment of the application, before training the initial regression model, the process of constructing the initial regression model is also included. The initial regression model can be an XGBoost regression model. The XGBoost regression model is an ensemble learner that combines several weak learners with regression trees as weak learners. The ensemble process of several weak learners can be represented by the following mathematical expression, formula (3): Formula (3); in, This represents the XGBoost regression model. For the k-th tree (weak classifier), Here are the parameters for the k-th decision tree, and K is the number of decision trees.

[0077] The mathematical expression for the Kth model is as follows: (4) Formula (4); Among them, the k-th tree Obtained from residual training, residual Initial promotion tree .

[0078] The parameters of the k-th decision tree are determined by minimizing empirical risk. , The mathematical expression is as follows: (5) Formula (5); The loss function for the regression class is: y is the true value. These are predicted values.

[0079] Substituting the loss function into the above formula (5), we obtain the objective function described in the following formula (6): Formula (6); The training process of the initial regression model is the process of finding the minimum value of the objective function.

[0080] After the model training is completed, the air conditioner filter life cycle prediction model can be tested based on the test set. In this embodiment, the air conditioner filter life cycle prediction model can be iterated at preset time intervals based on the same model training method to ensure the prediction accuracy of the air conditioner filter life cycle prediction model.

[0081] In this embodiment, since the health status of the replaced air conditioning filter is used as a label when training the initial regression model, and the health status of the replaced air conditioning filter is obtained based on the weight, air resistance and purification efficiency of the replaced air conditioning filter, the accuracy of the obtained air conditioning filter life prediction model can be improved by using the filter health status as a label to train the initial regression model.

[0082] The air conditioning filter usage cycle prediction method described in this application embodiment can be applied to a server to reduce the computational burden on the vehicle. The trained air conditioning filter usage cycle prediction model can be deployed in the cloud to predict the usage cycle of the vehicle's air conditioning filter.

[0083] Compared to current fixed maintenance rules, the air conditioning filter usage cycle prediction method provided in this application can intelligently extend the usage time of the air conditioning filter for users with short-cycle filter replacements based on the specific usage of the target vehicle, making full use of the filter's residual value. Alternatively, it can shorten the reminder cycle for users with long-cycle filter replacements, enabling personalized display of the remaining usage days of the air conditioning filter for each vehicle, which is beneficial for ensuring air quality and user health. Furthermore, this application embodiment does not require additional sensors, has no impact on vehicle weight or BOM cost, and the prediction results can be directly used for display on the vehicle or app, pushing maintenance reminder messages to users and improving the user experience.

[0084] Based on the same inventive concept, as an implementation of the above-mentioned method for predicting the service life of an air conditioning filter, this application also provides an air conditioning filter service life prediction device that performs the above-mentioned method embodiment. This device embodiment corresponds to the aforementioned method embodiment. For ease of reading, this device embodiment will not repeat the details of the aforementioned method embodiment one by one, but it should be clear that the air conditioning filter service life prediction device in this embodiment can correspondingly implement all the contents of the aforementioned method embodiment for predicting the service life of an air conditioning filter.

[0085] Figure 3 This is a schematic diagram of the structure of an air conditioning filter life prediction device provided in one embodiment of this application, as shown below. Figure 3 As shown, the air conditioning filter usage cycle prediction device 300 provided in this embodiment includes: The acquisition module 310 is used to acquire signal data of the target vehicle within a preset time period, weather data of the target vehicle during the driving within the preset time period, and filter after-sales maintenance data of the target vehicle. The preset time period is the time period from the start time of the current filter of the target vehicle to the current time. The prediction module 320 is used to input the signal data, the weather data, and the filter after-sales maintenance data into the air conditioning filter usage cycle prediction model, and use the air conditioning filter usage cycle prediction model to predict and output the filter health status of the current filter at the current time; wherein, the air conditioning filter usage cycle prediction model is obtained by training an initial regression model using the filter health status of replaced air conditioning filters in the training dataset as labels, and the filter health status of replaced air conditioning filters is calculated by the performance data matrix of replaced air conditioning filters, which is used to indicate the filtration performance of replaced air conditioning filters, and the performance data matrix of replaced air conditioning filters is a matrix composed of the weight, air resistance, and purification efficiency of replaced air conditioning filters; The conversion module 330 is used to convert the current filter element's filter element health status at the current moment into the predicted usage cycle of the current filter element.

[0086] As an optional implementation of this application, the conversion module 330 is specifically used to map the filter health status to the initial predicted usage period of the current filter at the current moment through linear interpolation, and to obtain the predicted usage periods corresponding to multiple dates before the current moment; and to use the average of the initial predicted usage period corresponding to the current moment and the predicted usage periods corresponding to multiple dates before the current moment as the predicted usage period of the air conditioning filter at the current moment.

[0087] Figure 4 This is a schematic diagram of the structure of an air conditioning filter usage cycle prediction device provided in one embodiment of this application. Figure 3 Based on the device shown, it also includes: The display module 410 is used to convert the current filter's health status at the current moment into the predicted usage period of the current filter, calculate the remaining usage days of the current filter based on the predicted usage period of the current filter at the current moment and the number of days already used, and send the remaining usage days to the vehicle for display.

[0088] As an optional implementation of this application, the display module 410 is used to calculate the time interval between the current time and the remaining usage days after sending the remaining usage days to the vehicle terminal for display; if the time interval is less than or equal to a preset time interval, a reminder message is displayed on the vehicle terminal, and the reminder message is used to remind the user to replace the air conditioning filter.

[0089] As an optional implementation of this application, the device further includes: a training module 420, used to train the initial regression model before acquiring the signal data of the target vehicle within a preset time period, the weather data of the target vehicle during the preset time period, and the filter after-sales maintenance data of the target vehicle. The training of the initial regression model includes: acquiring sample data for each of the multiple vehicles, wherein the sample data for each vehicle includes: signal data of the vehicle during the period of use after the replacement of the air conditioning filter, weather data of the vehicle during the period of use after the replacement of the air conditioning filter, after-sales maintenance data of the vehicle's filter, and the filter health of the vehicle's replaced air conditioning filter; constructing a training dataset by using the signal data, weather data, and after-sales maintenance data of the multiple vehicles during the period of use after the replacement of the air conditioning filter as a feature set, and using the filter health of the multiple vehicles' replaced air conditioning filters as a label; inputting the feature set in the training dataset into the initial regression model, adjusting the parameters of the initial regression model based on the predicted values ​​and labels output by the initial regression model until the training termination condition is met, thereby obtaining the air conditioning filter usage cycle prediction model.

[0090] As an optional implementation of this application, the acquisition module 310 is specifically used to acquire the filter health of the replaced air conditioning filter of each vehicle; wherein, for each vehicle, acquiring the filter health of the replaced air conditioning filter includes: collecting multiple replaced air conditioning filters of the vehicle, testing each collected replaced air conditioning filter to obtain the weight, air resistance, and remaining purification efficiency of each replaced air conditioning filter, and taking the difference between 1 and the remaining purification efficiency of each replaced air conditioning filter as the purification efficiency of the corresponding replaced air conditioning filter; constructing the performance data matrix by using the weight, air resistance, and purification efficiency of each replaced air conditioning filter as row elements; performing dimensionality reduction and weighting processing on the performance data matrix to obtain the residual filtration performance index of the replaced air conditioning filter of the vehicle; calculating the difference between 1 and the residual filtration performance index of the replaced air conditioning filter of the vehicle as the filter health of the replaced air conditioning filter of the vehicle.

[0091] As an optional implementation of this application, the acquisition module 310 is specifically used to normalize each row of the performance data matrix and calculate the covariance matrix of the normalized performance data matrix; calculate the eigenvalues ​​of the covariance matrix and the eigenvectors corresponding to each eigenvalue; arrange the eigenvectors corresponding to each eigenvalue from top to bottom according to the eigenvalues ​​in descending order to obtain the feature matrix corresponding to the performance data matrix; and use the eigenvalues ​​to weight each column of the matrix obtained by multiplying the feature matrix and the performance data matrix to obtain the residual filtration performance index of the replaced air conditioning filter of the vehicle.

[0092] As an optional implementation of this application, the training module 420 is specifically used to sort the signal data of each vehicle during the period of use of the replaced air conditioning filter, the weather data of the vehicle during the period of use of the replaced air conditioning filter, and the filter after-sales maintenance data of the vehicle to obtain a feature data sorting result; remove data with lower importance than a preset importance from the feature data sorting result to obtain filtered feature data; and construct a training dataset by using the filtered feature data of the multiple vehicles as a feature set and the filter health of the replaced air conditioning filters of the multiple vehicles as a label.

[0093] This embodiment of the device will not repeat the details of the aforementioned method embodiments one by one, but it should be clear that the air conditioning filter life prediction device in this embodiment can realize all the contents of the aforementioned air conditioning filter life prediction method embodiments.

[0094] In one embodiment, an electronic device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of any of the air conditioning filter life prediction methods described in the above method embodiments, or the processor executes the computer program to implement the steps of any of the air conditioning filter life prediction methods described in the above method embodiments.

[0095] For example, Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 5 As shown, the electronic device provided in this embodiment includes a memory 51 and a processor 52. The memory 51 is used to store computer programs; the processor 52 is used to execute the steps in the air conditioning filter usage cycle prediction method provided in the above method embodiment when the computer program is called, or, the processor 52 is used to execute the steps in the air conditioning filter usage cycle prediction method provided in the above method embodiment when the computer program is called. The implementation principle and technical effects are similar, and will not be described again here. Those skilled in the art will understand that... Figure 5The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0096] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps of any of the air conditioning filter life prediction methods described in the above method embodiments, or, when the computer program is executed by a processor, it implements the steps of any of the air conditioning filter life prediction methods described in the above method embodiments.

[0097] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static random access memory (SRAM) and dynamic random access memory (DRAM), etc.

[0098] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0099] For ease of explanation, the above description has been provided in conjunction with specific embodiments. However, the discussion in some embodiments above is not intended to be exhaustive or to limit the embodiments to the specific forms disclosed above. Various modifications and variations can be obtained based on the above teachings. The selection and description of the above embodiments are for the purpose of better explaining the principles and practical applications, thereby enabling those skilled in the art to better utilize the embodiments and various different variations of the embodiments suitable for specific application considerations.

Claims

1. A method for predicting the service life of an air conditioning filter, characterized in that, include: The system acquires signal data of the target vehicle within a preset time period, weather data of the target vehicle during the preset time period, and after-sales maintenance data of the target vehicle's filter element. The preset time period is the time period from the start time of the target vehicle's current filter element to the current time. The signal data, weather data, and filter after-sales maintenance data are input into the air conditioning filter usage cycle prediction model. The air conditioning filter usage cycle prediction model is used to predict and output the current filter health status at the current time. The air conditioning filter usage cycle prediction model is obtained by training an initial regression model using the filter health status of replaced air conditioning filters in the training dataset as labels. The filter health status of replaced air conditioning filters is calculated by the performance data matrix of replaced air conditioning filters, which is used to indicate the filtration performance of replaced air conditioning filters. The performance data matrix of replaced air conditioning filters is a matrix composed of the weight, air resistance, and purification efficiency of replaced air conditioning filters. The current filter element's health status at the current moment is converted into the predicted usage cycle of the current filter element.

2. The method according to claim 1, characterized in that, The step of converting the current filter's health status at the current moment into a predicted usage cycle for the current filter includes: The filter health status is mapped to the initial predicted usage period of the current filter at the current moment by linear interpolation, and the predicted usage periods corresponding to multiple dates before the current date are obtained respectively. The predicted usage period of the air conditioning filter at the current moment is calculated by taking the average of the initial predicted usage period corresponding to the current moment and the predicted usage periods corresponding to multiple dates prior to the current moment.

3. The method according to claim 1, characterized in that, After converting the current filter element's filter element health status at the current moment into the predicted usage cycle of the current filter element, the method further includes: The remaining usage days of the current filter element are calculated based on its predicted usage cycle and the number of days it has been used at the current moment, and then the remaining usage days are sent to the vehicle for display.

4. The method according to claim 3, characterized in that, After sending the remaining usage days to the vehicle for display, the method further includes: Calculate the time interval between the current moment and the remaining number of days of use; If the time interval is less than or equal to a preset time interval, a reminder message will be displayed on the vehicle side to remind the user to replace the air conditioning filter.

5. The method according to any one of claims 1-4, characterized in that, Before acquiring the signal data of the target vehicle within a preset time period, the weather data of the target vehicle during the preset time period, and the filter after-sales maintenance data of the target vehicle, the method further includes: training the initial regression model. Training the initial regression model includes: For multiple vehicles, sample data for each of the multiple vehicles is obtained, wherein the sample data for each vehicle includes: signal data of the vehicle during the period when the air conditioning filter has been replaced, weather data of the vehicle during the period when the air conditioning filter has been replaced, after-sales maintenance data of the vehicle's filter, and the filter health of the vehicle's replaced air conditioning filter. The signal data of the multiple vehicles during the period when the air conditioning filter was replaced, the weather data of the multiple vehicles during the period when the air conditioning filter was replaced, and the after-sales maintenance data of the multiple vehicles were used as feature sets, and the filter health of the replaced air conditioning filter of the multiple vehicles was used as a label to construct a training dataset. The feature set in the training dataset is input into the initial regression model. The parameters of the initial regression model are adjusted based on the predicted values ​​and labels output by the initial regression model until the training termination condition is met, thus obtaining the air conditioner filter usage cycle prediction model.

6. The method according to claim 5, characterized in that, The step of obtaining sample data for each of the plurality of vehicles includes: Obtain the filter health status of the replaced air conditioning filter for each vehicle; Specifically, for each vehicle, the health status of the replaced air conditioning filter is obtained, including: Multiple replaced air conditioning filters of the vehicle are collected, and each replaced air conditioning filter is tested to obtain the weight, wind resistance and remaining purification efficiency of each replaced air conditioning filter. The difference between 1 and the remaining purification efficiency of each replaced air conditioning filter is taken as the purification efficiency of the corresponding replaced air conditioning filter. The performance data matrix is ​​constructed by using the weight, air resistance, and purification efficiency of each replaced air conditioning filter as row elements. The performance data matrix is ​​subjected to dimensionality reduction and weighting to obtain the residual filtration performance index of the replaced air conditioning filter of the vehicle. The difference between 1 and the residual filtration performance index of the replaced air conditioning filter of the vehicle is used as the filter health of the replaced air conditioning filter of the vehicle.

7. The method according to claim 6, characterized in that, The step of performing dimensionality reduction and weighting on the performance data matrix to obtain the residual filtration performance index of the replaced air conditioning filter of the vehicle includes: Normalize each element of the performance data matrix and calculate the covariance matrix of the normalized performance data matrix; Calculate the eigenvalues ​​of the covariance matrix and the eigenvectors corresponding to each eigenvalue; Arrange the eigenvectors corresponding to each eigenvalue from top to bottom according to the eigenvalues ​​in descending order to obtain the eigenma matrix corresponding to the performance data matrix; By using the eigenvalues ​​to weight each column of the matrix obtained by multiplying the feature matrix and the performance data matrix, the residual filtration performance index of the replaced air conditioning filter of the vehicle is obtained.

8. The method according to claim 5, characterized in that, The training dataset is constructed by using signal data of the multiple vehicles during the period when the air conditioning filters have been replaced, weather data of the multiple vehicles during the period when the air conditioning filters have been replaced, and after-sales maintenance data of the multiple vehicles' filters as feature sets, and using the filter health status of the replaced air conditioning filters of the multiple vehicles as labels. The dataset includes: The signal data of each vehicle during the period when the air conditioning filter has been replaced, the weather data of the vehicle during the period when the air conditioning filter has been replaced, and the after-sales maintenance data of the filter of the vehicle are ranked by importance to obtain the feature data ranking result. Data with importance lower than a preset importance in the sorted feature data are removed to obtain the filtered feature data; The selected feature data of the multiple vehicles are used as the feature set, and the health status of the replaced air conditioning filters of the multiple vehicles is used as the label to construct a training dataset.

9. An air conditioning filter life prediction device, characterized in that, include: The acquisition module is used to acquire signal data of the target vehicle within a preset time period, weather data of the target vehicle during the driving within the preset time period, and filter after-sales maintenance data of the target vehicle. The preset time period is the time period from the start time of the current filter of the target vehicle to the current time. The prediction module is used to input the signal data, the weather data, and the filter after-sales maintenance data into the air conditioning filter usage cycle prediction model, and use the air conditioning filter usage cycle prediction model to predict and output the filter health status of the current filter at the current time; wherein, the air conditioning filter usage cycle prediction model is obtained by training an initial regression model using the filter health status of replaced air conditioning filters in the training dataset as labels, and the filter health status of replaced air conditioning filters is calculated by the performance data matrix of replaced air conditioning filters, which is used to indicate the filtration performance of replaced air conditioning filters, and the performance data matrix of replaced air conditioning filters is a matrix composed of the weight, air resistance, and purification efficiency of replaced air conditioning filters; The conversion module is used to convert the current filter's health status at the current moment into the predicted usage cycle of the current filter.

10. An electronic device, comprising: A memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the air conditioning filter usage cycle prediction method according to any one of claims 1 to 8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the air conditioning filter usage cycle prediction method according to any one of claims 1 to 8.

12. A vehicle, characterized in that, The vehicle is equipped with the air conditioning filter usage cycle prediction device of claim 9, or the electronic device of claim 10, or the storage medium of claim 11.

Images