A vehicle-mounted air conditioner filter element life prediction system

Abstract

The application belongs to the technical field of automobile parts, and provides a vehicle-mounted air conditioner filter core life prediction system, which comprises the following steps: collecting the performance index vector of the vehicle-mounted air conditioner filter core, judging whether the vehicle-mounted air conditioner filter core is in a stable attenuation period, if yes, marking the stable attenuation period, and triggering the vehicle-mounted air conditioner filter core life prediction, collecting the working information of the vehicle-mounted air conditioner in the historical period, dividing the working condition types of the historical period into time sequence working condition sequences, periodically analyzing the time sequence working condition sequences, if there is periodicity, calculating the working condition period, and judging whether to trigger the dynamic working condition attenuation adaptation, if not, triggering the average attenuation analysis, based on the performance index vector of each time, constructing an average performance attenuation model, if the dynamic working condition attenuation adaptation is triggered, constructing a working condition performance attenuation rate change model, and using the average performance attenuation model or the working condition performance attenuation rate change model to predict the remaining life of the vehicle-mounted air conditioner filter core.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of automobile accessories, in particular to a system for predicting the service life of a vehicle-mounted air conditioner filter element. BACKGROUND

[0002] As a core component for ensuring the air quality inside the vehicle, the vehicle-mounted air conditioner filter element mainly functions to filter particulate matter, pollen, peculiar smell and some harmful gases in the air entering the vehicle, directly affecting the respiratory health of the people inside the vehicle and the driving experience. With the passage of time, the filter element will gradually deteriorate due to the attachment of particulate matter and the aging of fibers, resulting in a decrease in filtering efficiency and an increase in air resistance. When the performance deteriorates to a critical value, it needs to be replaced in time, otherwise it may lead to problems such as insufficient air volume of the air conditioner and deterioration of the air quality inside the vehicle.

[0003] However, the existing life management mode of the vehicle-mounted air conditioner filter element has many technical problems. The discovery of these problems is closely related to the development of the industry and the upgrading of user needs. In the early stage, the industry generally adopted a fixed cycle replacement mode. However, it was found in practice that the actual deterioration rate of the filter element was significantly affected by external environment, usage habits and other factors. For example, the deterioration rate of the filter element of a vehicle that has been driving in a serious smog area for a long time is much faster than that of a vehicle that has been used in a clean environment. Frequent opening of the large air volume mode also accelerates the clogging of the filter element. The fixed cycle cannot adapt to individual differences, often leading to excessive replacement or replacement lag.

[0004] With the increasing attention of users to the health inside the vehicle, some technologies attempt to judge the life through subjective feelings or a single parameter, such as relying on the decrease of air volume and the obvious peculiar smell perceived by the user, or only monitoring the pressure difference between the two sides of the filter element to judge the clogging degree. However, the former has a serious lag and is often discovered after the filter element fails. The latter is one-sided and only reflects the clogging state of the filter element, which cannot directly correlate with the filtering efficiency, resulting in a high risk of misjudgment.

[0005] Further research shows that the performance deterioration of the filter element is a process of coordinated change of multiple parameters, and there is a strong correlation between the working conditions of the vehicle air conditioner. In the existing technology, there is a lack of real-time monitoring and quantitative analysis of core performance indicators such as filtering efficiency and resistance. Moreover, a dynamic correlation model between working conditions and deterioration rate has not been established, resulting in low accuracy of life prediction. In addition, the existing prediction methods are mostly static estimates, which cannot dynamically adjust the prediction results according to the real-time performance state of the filter element and the changes in working conditions, making it difficult to meet the needs of users for accurate life warning. There is a lack of dynamic adjustment mechanism, which cannot update the life prediction according to the real-time state, and it is easy to cause the user to miss the best replacement time.

[0006] In view of the above problems, the present application provides a system for predicting the service life of a vehicle-mounted air conditioner filter element. SUMMARY

[0007] To make up for the deficiencies of the prior art, solve at least one technical problem proposed in the background art.

[0008] The technical scheme adopted by the present application to solve the technical problem is: a vehicle-mounted air conditioner filter core life prediction system, comprising:

[0009] The performance analysis module collects the performance index vector of the vehicle-mounted air conditioner filter core, judges in real time whether the vehicle-mounted air conditioner filter core is in a stable decay period, if yes, marks the stable decay period, and triggers the vehicle-mounted air conditioner filter core life prediction;

[0010] The working condition analysis module collects the working information of the vehicle-mounted air conditioner in the historical period if the vehicle-mounted air conditioner filter core life prediction is triggered, divides the working condition type for each time and constructs the time sequence working condition sequence, periodically analyzes the time sequence working condition sequence, calculates the working condition period if there is periodicity, and judges whether to trigger the dynamic working condition decay adaptation, if not, triggers the average decay analysis;

[0011] The model construction module constructs the average performance decay model if the average decay analysis is triggered based on the performance index vector of each time, and constructs the working condition performance decay rate change model based on the time sequence periodic working condition sequence group if the dynamic working condition decay adaptation is triggered in the stable decay period.

[0012] The prediction and deduction module predicts the remaining life of the vehicle-mounted air conditioner filter core using the obtained average performance decay model or working condition performance decay rate change model.

[0013] The judgment method of whether the vehicle-mounted air conditioner filter core is in a stable decay period is:

[0014] The judgment result of whether the vehicle-mounted air conditioner filter core is in a stable period is obtained, if the vehicle-mounted air conditioner filter core is not in a stable period, the Euclidean distance between the performance index vector of the vehicle-mounted air conditioner filter core at the current time and the previous time is calculated, which is marked as the adjacent decay distance, and the Euclidean distance between the performance index vector of the vehicle-mounted air conditioner filter core and the preset performance index vector bottom line is calculated, which is marked as the decay endpoint distance.

[0015] If the adjacent decay distance at the current time is less than or equal to the preset instantaneous change standard, and the decay endpoint distance at the current time is greater than or equal to the preset distance standard, it is judged that the vehicle-mounted air conditioner filter core is in a stable decay period.

[0016] The judgment method of whether the vehicle-mounted air conditioner filter core is in a stable period is:

[0017] Obtaining a performance index vector of the vehicle-mounted air conditioner filter element at a starting time point when the vehicle-mounted air conditioner filter element is put into use, marking as an initial performance index vector, calculating the Euclidean distance between the performance index vector of the vehicle-mounted air conditioner filter element and the initial performance index vector, if less than a preset distance standard, judging that the vehicle-mounted air conditioner filter element is in a stable period;

[0018] The manner for dividing the working condition types at the time points is:

[0019] Obtaining working information of the vehicle-mounted air conditioner in a historical period, the working information including a wind volume gear and an external environment parameter, performing normalization processing and integration on the working information in the historical period respectively to obtain a normalized working information vector at each time point;

[0020] Dividing the time points with the wind volume gear being 0 into the same working condition type, dividing the time points with the wind volume gear not being 0 into working condition types by using a K-means clustering algorithm, determining the number of clusters for the time points with the wind volume gear not being 0 in the historical period by using an elbow method and performing iterative clustering until convergence is determined, and then ending the clustering to obtain a plurality of temporary clusters, each temporary cluster corresponding to a working condition type, and dividing the time points corresponding to the normalized working information vectors belonging to the same temporary cluster into the same working condition type;

[0021] The manner for constructing the time sequence working condition sequence is:

[0022] Assigning a unique working condition digital code to each working condition type, intercepting a sliding analysis window with a fixed time length from the current time point as an end point, obtaining the working condition digital code corresponding to the working condition type of each time point in the sliding analysis window, and integrating according to the time sequence to obtain the time sequence working condition sequence;

[0023] The manner for judging whether to trigger the dynamic working condition attenuation adaptation is:

[0024] Performing regular analysis on the time sequence working condition sequence by using an autocorrelation function, analyzing whether the time sequence working condition sequence has periodicity, if the time sequence working condition sequence has significant periodicity, obtaining a working condition period according to the autocorrelation function, and if there are not less than two complete working condition periods in the stable attenuation period, judging to trigger the dynamic working condition attenuation adaptation;

[0025] The manner for constructing the average performance attenuation model is:

[0026] Respectively obtaining a performance index vector at a starting time point of the stable attenuation period and a performance index vector at the current time point, the performance index vector including a filtration efficiency and a resistance ratio, calculating an average attenuation rate of the filtration efficiency and an average growth rate of the resistance ratio in the stable attenuation period, combining the performance index vector at the current time point, constructing a model of the filtration efficiency and the resistance ratio changing with time, and integrating to obtain the average performance attenuation model;

[0027] The time sequence cycle working condition sequence group is obtained by:

[0028] The time points of continuous same working condition types are merged into a working condition period, the same working condition type is the working condition type of the working condition period, the performance index vector of the start point and the end point of each working condition period in the stable decay period is obtained, and the average decay rate of the filtration efficiency and the average growth rate of the resistance ratio in the working condition period are calculated;

[0029] All working condition periods in a working condition cycle are obtained, and the time sequence is integrated and sequentially marked with a cycle number, to obtain a time sequence cycle working condition sequence. All time sequence cycle working condition sequences in the stable decay period are integrated according to the time sequence and sequentially marked with a sequence number, to obtain a time sequence cycle working condition sequence group. The cycle number and the sequence number are both counted as integers starting from 1;

[0030] The working condition performance decay rate change model is constructed in the following manner:

[0031] In the time sequence cycle working condition sequence group, all working condition periods with the same cycle number are integrated according to the time sequence, to obtain a time sequence working condition period sequence corresponding to the cycle number;

[0032] In each time sequence working condition period sequence, the sequence number corresponding to the working condition period is taken as the independent variable, the average decay rate of the filtration efficiency and the average growth rate of the resistance ratio in the working condition period are taken as the dependent variables, a linear relationship is fitted by the least square method, the filtration efficiency decay rate change model and the resistance ratio growth rate change model of each cycle number are obtained, and are integrated to obtain the working condition performance decay rate change model;

[0033] The working condition performance decay rate change model is used to predict the remaining life of the vehicle-mounted air conditioner filter element in the following manner:

[0034] A deduction period is set with the current time as the starting point. If the current time is not the end point of the working condition cycle, the current working condition cycle is predicted and completed in the deduction period. The working condition performance decay rate change model is used to sequentially calculate the Euclidean distance between the performance index vector at the end point of each working condition period in the deduction period and the preset performance index vector bottom line, which is marked as the predicted end-of-life distance. If the predicted end-of-life distance is less than the preset distance standard, the deduction is stopped.

[0035] If the predicted end-of-life distance of the current working condition cycle end point is still greater than or equal to the preset distance standard, the next working condition cycle is continued to be deduced in the deduction period until there is a working condition period that makes the deduction stop, which is marked as the termination working condition period. The predicted end-of-life distance of the time point in the termination working condition period is sequentially predicted and calculated according to the time sequence. If the predicted end-of-life distance of any time point is less than the preset distance standard, the time point is judged as the end-of-life of the vehicle-mounted air conditioner filter element, and the subsequent time points are not calculated.

[0036] The interval between the current time and the end of life is calculated to obtain the remaining life of the vehicle-mounted air conditioner filter element.

[0037] The beneficial effects of the present application are as follows:

[0038] 1、The present application can timely determine whether the vehicle-mounted air conditioner filter element enters the stable decay period and mark the period by accurately collecting the performance index vector of the vehicle-mounted air conditioner filter element, thereby providing an accurate basis for subsequent analysis. By collecting historical working information, dividing the working condition types, constructing the time sequence working condition sequence and performing periodic analysis, the use environment characteristics of the filter element can be comprehensively mastered. According to different trigger conditions, the average or working condition performance decay model can be constructed, which can be fitted to the actual use situation, effectively improving the accuracy of the prediction of the remaining life of the vehicle-mounted air conditioner filter element, and providing a scientific basis for filter element replacement.

[0039] 2、After determining that the stable decay period is entered, the historical working conditions are analyzed in detail. Whether the average decay analysis is triggered to construct the average performance decay model or the dynamic working condition decay adaptation is triggered to construct the working condition performance decay rate change model, various use scenarios can be fully considered. This flexible and accurate modeling method makes the prediction result more in line with the actual situation, helps users to plan filter element replacement in advance, reduces the influence of poor filter element performance on indoor air quality and air conditioning system, and improves the driving experience. BRIEF DESCRIPTION OF DRAWINGS

[0040] The present application will be further described below in conjunction with the drawings.

[0041] Figure 1 is a module architecture diagram of a vehicle-mounted air conditioner filter element life prediction system according to an embodiment of the present application;

[0042] Figure 2 is a prediction step flowchart of the remaining life of the vehicle-mounted air conditioner filter element in the vehicle-mounted air conditioner filter element life prediction system according to an embodiment of the present application. DETAILED DESCRIPTION

[0043] In order to make the technical means, creative features, purposes and effects achieved by the present application easy to understand, the present application will be further described below in conjunction with specific embodiments.

[0044] Embodiment 1

[0045] Please refer to Figure 1 The vehicle-mounted air conditioner filter element life prediction system according to an embodiment of the present application includes the following steps:

[0046] Performance analysis module: collect the performance index vector of the vehicle-mounted air conditioner filter element, determine whether the vehicle-mounted air conditioner filter element is in the stable decay period, if yes, mark the stable decay period, and trigger the vehicle-mounted air conditioner filter element life prediction;

[0047] A sensor group consisting of a particle counter and a pressure sensor is installed in the vehicle air conditioner. The particle counter is installed at the air inlet and outlet of the vehicle air conditioner filter, and the pressure sensor is installed on both sides of the vehicle air conditioner filter. The installed sensor group collects key performance indicators of the vehicle air conditioner filter in real time, including filtration efficiency and resistance ratio. The key performance indicators are integrated into a performance indicator vector at each moment.

[0048] The filtration efficiency η reflects the filter element's ability to filter particulate matter in the air. It is calculated by using a particle counter to simultaneously detect the particle concentration at the air inlet and outlet of the vehicle air conditioning filter element. The formula is:

[0049] η=(particle concentration at the air inlet - particle concentration at the air outlet) / particle concentration at the air inlet × 100%;

[0050] Among them, the resistance ratio Reflects the difficulty of air passing through the filter element. It is calculated by detecting the air pressure values ​​on both sides of the vehicle air conditioning filter element through the pressure sensor. The air pressure values ​​on both sides of the vehicle air conditioning filter element include the air pressure value on the air inlet side and the air pressure value on the air outlet side. The formula is:

[0051] = (air inlet pressure value - air outlet pressure value) / air inlet pressure value × 100%;

[0052] Obtain the performance index vector of the vehicle air conditioning filter at the starting time when the vehicle air conditioning filter is put into use, mark it as the initial performance index vector, and calculate the Euclidean distance between the performance index vector of the vehicle air conditioning filter at the current time and the initial performance index vector. If the calculated Euclidean distance is less than a preset distance standard, it is determined that the vehicle air conditioning filter has not experienced significant attenuation at the current time and is currently in a stable period of the vehicle air conditioning filter.

[0053] If the current moment is not in the stable period of the vehicle air conditioning filter, calculate the Euclidean distance between the performance index vector of the vehicle air conditioning filter at the current moment and the previous moment, and mark it as the adjacent attenuation distance at the current moment. Calculate the Euclidean distance between the performance index vector of the vehicle air conditioning filter at the current moment and the preset performance index vector bottom line, and mark it as the attenuation end distance at the current moment.

[0054] If the adjacent attenuation distance at the current moment is greater than the preset instantaneous change standard, or the attenuation end distance at the current moment is less than the preset distance standard, it is judged that the vehicle air conditioning filter element has experienced cliff-like attenuation, and the vehicle air conditioning filter element has reached the end of its life at the current moment. Otherwise, it is judged that the current moment is in a stable attenuation period, and the stable attenuation period will be marked from the moment of the first stable attenuation period to the current moment as the end point, triggering the vehicle air conditioning filter element life prediction;

[0055] If the vehicle-mounted air conditioner filter core reaches the end of life, a filter core replacement reminder is generated and sent to the user terminal;

[0056] It should be noted that the function of this module is to collect the filtration efficiency and resistance ratio of the vehicle-mounted air conditioner filter core in real time through the sensor group, construct a performance index vector, quantify the performance change based on the Euclidean distance, determine whether the vehicle-mounted air conditioner filter core is in the stable period, stable decay period or end of life, and trigger life prediction or directly generate a replacement reminder, thereby realizing real-time monitoring and accurate state division of the performance of the vehicle-mounted air conditioner filter core, avoiding premature replacement and wasting costs, or replacing too late and affecting the air conditioning effect and health, improving the use efficiency of the filter core, introducing the difference of the Euclidean distance quantified performance index vector, breaking through the limitations of traditional single index life prediction, and realizing dynamic identification of multi-dimensional performance decay;

[0057] Working condition analysis module: if the vehicle-mounted air conditioner filter core life prediction is triggered, the working information of the vehicle-mounted air conditioner in the historical period is collected, the working condition type is divided for each time in the historical period, the time sequence is constructed based on the working condition type, the periodicity of the time sequence is analyzed, if there is periodicity, the working condition cycle is calculated, and it is judged whether to trigger dynamic working condition decay adaptation, if not, the average decay analysis is triggered;

[0058] If the vehicle-mounted air conditioner filter core life prediction is triggered, the starting time of the vehicle-mounted air conditioner filter core is obtained, the period between the starting time and the current time is marked as the historical period, the historical working information of the vehicle-mounted air conditioner is recorded and collected through the vehicle-mounted system, the historical working information includes the working information of the vehicle-mounted air conditioner at each time in the historical period, the working information includes the air volume gear and the external environment parameters;

[0059] Among them, the air volume gear is obtained through the electronic operation record of the vehicle-mounted air conditioner control panel, if the vehicle-mounted air conditioner is not turned on, the air volume gear is 0, the external environment parameters include the inlet particle concentration, the environment temperature and the environment humidity, the inlet particle concentration is obtained by the particle counter installed in the inlet of the vehicle-mounted air conditioner filter core, the environment temperature is collected by the temperature sensor installed outside the vehicle, and the environment humidity is collected by the vehicle-mounted humidity sensor;

[0060] In the historical period, each type of working information is normalized, and the normalized working information is integrated at each time to obtain the normalized working information vector of each time;

[0061] For the time when the air volume gear is not 0, the K-means clustering algorithm is used to divide the working condition type for each time;

[0062] Specifically, the number of clusters is determined for the time when the air volume gear is not 0 in the historical period by the elbow method, the normalized working information vectors of the number of clusters are selected as cluster centers, the Euclidean distance between each normalized working information vector and the cluster center is calculated, each normalized working information vector is assigned to the cluster where the Euclidean distance is closest to the cluster center, forming the number of temporary clusters, for each temporary cluster, the average of each component of all normalized working information vectors in the temporary cluster is calculated as a new cluster center, the difference between the new cluster center and the last cluster center is compared, if the change of all cluster centers is less than the set convergence threshold, it is determined that the convergence is reached, and the clustering ends, otherwise, each normalized working information vector is re-assigned to the cluster where the Euclidean distance is closest to the cluster center and the cluster center is updated until convergence is reached.

[0063] After clustering, the number of temporary clusters is obtained, each temporary cluster corresponds to a type of working condition, the time corresponding to the normalized working information vector belonging to the same temporary cluster is divided into the same working condition type, the time when the air volume gear is 0 is divided into the same working condition type, and a unique working condition digital code is assigned to each type of working condition.

[0064] A sliding analysis window with a fixed length is intercepted with the current time as the end point, the sliding analysis window slides with the change of the current time, and the working condition digital code corresponding to each time is integrated according to the time sequence to obtain a time sequence working condition sequence.

[0065] The autocorrelation function is used to analyze the regularity of the time sequence working condition sequence, whether there is periodicity in the sliding analysis window for each working condition type, the formula of the autocorrelation function is:

[0066] ;

[0067] Among them, denotes the autocorrelation coefficient, denotes the lag value, denotes the working condition digital code corresponding to the tth time in the time sequence working condition sequence, denotes the mean value of the time sequence working condition sequence, and n denotes the length of the time sequence working condition sequence.

[0068] The autocorrelation coefficients of different lag values are calculated , wherein the lag value range is 1 to n / 2, if there is a lag value , such that is greater than the preset autocorrelation coefficient standard, and in the lag value range, the autocorrelation coefficients at the integer multiple lag values are all greater than the preset autocorrelation coefficient bottom line, it is determined that the time sequence working condition sequence has a significant periodicity, and the working condition period , wherein, denotes the interval duration between adjacent time instants;

[0069] If the time series of working conditions has significant periodicity and there are at least two complete working condition cycles in the stable decay period, triggering dynamic working condition decay adaptation, otherwise triggering average decay analysis;

[0070] It should be noted that the function of this module is to collect the historical working information of the vehicle-mounted air conditioner, divide the working condition types through clustering algorithm, analyze the periodicity of the working condition using autocorrelation function, trigger two prediction modes of average decay analysis or dynamic working condition decay adaptation according to the periodic characteristics, consider the influence of actual working condition on filter decay, avoid prediction deviation caused by working condition difference, lay a foundation for subsequent accurate prediction, convert complex working conditions into discrete code sequences for time series analysis, combine clustering algorithm with autocorrelation function, classify working conditions first and then analyze their periodicity, adapt the prediction mode to the actual use rule, and break through the traditional prediction idea of ignoring working condition difference;

[0071] Model construction module: based on the performance index vector at each time instant, if average decay analysis is triggered, an average performance decay model is constructed, and if dynamic working condition decay adaptation is triggered, the working condition period is marked according to the working condition type at each time instant, the time series of periodic working conditions is integrated in the stable decay period, and a working condition performance decay rate change model is constructed based on the time series of periodic working conditions;

[0072] If average decay analysis is triggered, the performance index vector at the start of the stable decay period and the performance index vector at the current time instant are obtained respectively;

[0073] For the performance index vector, the average decay rate of the filter efficiency included in the performance index vector in the stable decay period is calculated , and the average growth rate of the resistance ratio included in the performance index vector , the formula is:

[0074] ;

[0075] ;

[0076] wherein, denotes the duration of the stable decay period, and denote the filter efficiency and resistance ratio included in the performance index vector at the start of the stable decay period, and denote the filter efficiency and resistance ratio included in the performance index vector at the current time instant;

[0077] Based on the performance index vector at the current time, the average decay rate of the filtration efficiency and the average growth rate of the resistance ratio in the stable decay period are combined to construct a model of the filtration efficiency and the resistance ratio changing with time, and an average performance decay model is obtained by integration;

[0078] If the dynamic condition decay adaptation is triggered, the same condition type time points in succession are combined into a condition period, and the same condition type is the condition type of the condition period;

[0079] Based on the performance index vector at the start and end of each condition period in the stable decay period, the average decay rate of the filtration efficiency and the average growth rate of the resistance ratio in the condition period are calculated;

[0080] All condition periods in a condition cycle are obtained, integrated according to the time sequence, and sequentially labeled with a cycle index i to obtain a time sequence condition sequence, and all time sequence condition sequences in the stable decay cycle are integrated according to the time sequence and sequentially labeled with a sequence index j to obtain a time sequence condition sequence group, and the cycle index i and the sequence index j are both counted from 1;

[0081] In the time sequence condition sequence group, all condition periods with the same cycle index are integrated according to the time sequence to obtain a time sequence condition period sequence corresponding to the cycle index;

[0082] For any time sequence condition period sequence, in the time sequence condition period sequence, the sequence index corresponding to the condition period is taken as the independent variable, and the average decay rate of the filtration efficiency and the average growth rate of the resistance ratio in the condition period are taken as the dependent variables, a linear relationship is fitted by the least square method, and a filtration efficiency decay rate change model and a resistance ratio growth rate change model corresponding to the cycle index are constructed;

[0083] All filtration efficiency decay rate change models and resistance ratio growth rate change models corresponding to the cycle index are summarized and integrated to obtain a condition performance decay rate change model;

[0084] It should be noted that the role of this module is to construct the corresponding attenuation model according to the working condition analysis result, if the average attenuation analysis is triggered, the average attenuation rate of the filtration efficiency and the average growth rate of the resistance ratio in the stable attenuation period are calculated, the average performance attenuation model is constructed, if the dynamic working condition attenuation adaptation is triggered, the attenuation rate change model in different periods is fitted according to the working condition period, the working condition performance attenuation rate change model is formed, the average performance attenuation model is suitable for no significant periodic working condition, the working condition performance attenuation rate change model is suitable for periodic working condition, which is highly matched with the actual attenuation law, and the accuracy of subsequent prediction is improved, in dynamic working condition, different working condition periods in the period are modeled respectively, the linear relationship between the attenuation rate and the cycle sequence is fitted through the least square method, the dynamic influence of working condition change on the attenuation rate is captured, and the defect that the traditional average model cannot reflect the working condition fluctuation is overcome;

[0085] Prediction and deduction module: using the average performance attenuation model or the working condition performance attenuation rate change model, the remaining life of the vehicle air conditioner filter element is predicted;

[0086] Specifically, if the average performance attenuation model is obtained, the performance index vector of the current time is obtained, and the performance index vector bottom line of the vehicle air conditioner filter element is obtained. The performance index vector includes filtration efficiency and resistance ratio, and the performance index vector bottom line includes filtration efficiency threshold and resistance ratio threshold. The average performance attenuation model is used to calculate the required time length of the filtration efficiency and the resistance ratio of the vehicle air conditioner filter element to reach the filtration efficiency threshold and the resistance ratio threshold respectively. The two required time lengths obtained by comparison are compared, and the minimum required time length is selected as the remaining life of the vehicle air conditioner filter element.

[0087] If the working condition performance attenuation rate change model is obtained, a deduction period is set from the current time as the starting point. If the current time is the end point of a working condition cycle, it is judged that the current working condition cycle is complete, otherwise it is judged that the current working condition cycle is incomplete.

[0088] If the current working condition cycle is incomplete, the current working condition cycle is predicted and completed in the deduction period. The working condition performance attenuation rate change model is used to predict and calculate the average attenuation rate of the filtration efficiency and the average growth rate of the resistance ratio of each working condition period in the current working condition cycle based on the sequence number corresponding to the current working condition cycle. The performance index vector of each working condition period end point in the deduction period is calculated in sequence combined with the performance index vector of the current time, and the Euclidean distance between the performance index vector and the performance index vector bottom line is calculated, which is marked as the predicted life end point distance. If the predicted life end point distance is less than the preset distance standard, the deduction is stopped.

[0089] If the predicted end-of-life distance of the current operating condition cycle end is still greater than or equal to the preset distance standard, the next operating condition cycle is continued to be deduced from the current operating condition cycle end as the starting point within the deduction period, marked as a deduction operating condition cycle, the time sequence cycle operating condition sequence of the deduction operating condition cycle is arranged and incorporated into the time sequence cycle operating condition sequence group, and the sequence number is sequentially marked as N+1, where N is the sequence number of the time sequence cycle operating condition sequence of the current operating condition cycle;

[0090] Similarly, the average decay rate of the filtration efficiency and the average growth rate of the resistance ratio in each operating condition period in the deduction operating condition cycle are calculated based on the sequence number prediction, and the performance index vector at the end of each operating condition period in the deduction operating condition cycle is sequentially calculated combined with the performance index vector at the end of the current operating condition cycle, and the predicted end-of-life distance at the end of the operating condition period is calculated. If the predicted end-of-life distance is less than the preset distance standard, the deduction is stopped. If the predicted end-of-life distance of the deduction operating condition cycle end is still greater than or equal to the preset distance standard, the deduction operation is repeated until the deduction is stopped.

[0091] When the deduction is stopped, the operating condition period that causes the deduction to stop is determined, marked as a termination operating condition period, the performance index vector at the end of the termination operating condition period is calculated based on the performance index vector at the end of the previous operating condition period in the termination operating condition period, combined with the average decay rate of the filtration efficiency and the average growth rate of the resistance ratio in the termination operating condition period, and the performance index vector at the time is calculated according to the time sequence in sequence prediction, and the predicted end-of-life distance at the time is calculated. If the predicted end-of-life distance at any time is less than the preset distance standard, the time is judged as the end-of-life of the vehicle air conditioner filter element, and the subsequent time is no longer calculated.

[0092] The interval length between the current time and the end-of-life is calculated to obtain the remaining life of the vehicle air conditioner filter element;

[0093] Based on the remaining life of the vehicle air conditioner filter element, the predicted replacement time of the vehicle air conditioner filter element is generated and sent to the user terminal;

[0094] It should be noted that the function of this module is to use the average performance decay model or the operating condition performance decay rate change model to predict the remaining life, the average performance decay model predicts the life by calculating the shortest time when the performance index reaches the threshold, the operating condition performance decay rate change model accurately judges the end-of-life to a specific time by completing the incomplete cycle and continuously deducing the subsequent cycle, and finally generates the replacement time and pushes it to the user, providing accurate remaining life and replacement time for the user, facilitating advance planning and maintenance, and ensuring the filtering effect of the vehicle air conditioner and the air quality inside the vehicle. Based on the cycle completion and continuous deduction mechanism of the operating condition performance decay rate change model, the end-of-life can be accurately predicted to a specific time within the operating condition period, which realizes higher precision time sequence prediction and is more suitable for actual use scenarios compared with the traditional estimation of approximate time length.

[0095] The technical scheme of the embodiment of the present application is: collecting a performance index vector of a vehicle-mounted air conditioner filter element, judging whether the vehicle-mounted air conditioner filter element is in a stable attenuation period, if yes, marking the stable attenuation period, and triggering vehicle-mounted air conditioner filter element life prediction, if triggering the vehicle-mounted air conditioner filter element life prediction, collecting working information of the vehicle-mounted air conditioner in a historical period, dividing working condition types for time points in the historical period, constructing a time sequence working condition sequence based on the working condition types, periodically analyzing the time sequence working condition sequence, if there is periodicity, calculating a working condition period, and judging whether to trigger dynamic working condition attenuation adaptation, if not, triggering average attenuation analysis, based on the performance index vector of each time point, if triggering the average attenuation analysis, constructing an average performance attenuation model, if triggering the dynamic working condition attenuation adaptation, marking a working condition period according to the working condition type of the time point, integrating time sequence periodic working condition sequence groups in the stable attenuation period in combination with the working condition period, constructing a working condition performance attenuation rate change model based on the time sequence periodic working condition sequence groups, and predicting the remaining life of the vehicle-mounted air conditioner filter element by using the average performance attenuation model or the working condition performance attenuation rate change model.

[0096] The basic principles, main features and advantages of the present application are shown and described above. It should be understood by those skilled in the art that the present application is not limited by the above embodiments, and the above embodiments and descriptions in the specification are only to illustrate the principles of the present application. Without departing from the spirit and scope of the present application, various changes and improvements can be made to the present application, and these changes and improvements all fall within the scope of the present application. The scope of protection of the present application is defined by the appended claims and their equivalents.

Claims

1. A system for predicting the life of a vehicle air conditioning filter, characterized by: include: Performance Analysis Module: This module collects the performance indicator vectors of the vehicle air conditioning filter and determines in real time whether the filter is in a stable attenuation period. If so, it marks the stable attenuation period and triggers the life prediction of the filter. Working condition analysis module: If the vehicle air conditioning filter life prediction is triggered, the vehicle air conditioning operating information within the historical period is collected, the working condition type is divided into time series, and a time series working condition sequence is constructed. The time series working condition sequence is analyzed for periodicity. If periodicity exists, the working condition period is calculated, and it is determined whether to trigger dynamic working condition attenuation adaptation. If not, the average attenuation analysis is triggered. The method of dividing the working condition types at each moment is as follows: Obtain the operating information of the vehicle air conditioner during the historical period. The operating information includes the air volume level and external environmental parameters. Normalize and integrate the operating information during the historical period to obtain a normalized operating information vector at each moment. The moments when the air volume level is 0 are divided into the same working condition type. The K-means clustering algorithm is used to divide the working condition type for the moments when the air volume level is not 0. The elbow method is used to determine the number of clusters for the moments when the air volume level is not 0 in the historical period and clustering is performed iteratively until convergence is determined. The number of temporary clusters is obtained, and each temporary cluster corresponds to a type of working condition. The moments corresponding to the normalized working information vector belonging to the same temporary cluster are divided into the same working condition type. The construction method of the time sequence operating condition sequence is: Assign a unique working condition digital code to each working condition type. Take the current moment as the end point and intercept a sliding analysis window of constant length. Obtain the working condition digital code corresponding to the working condition type at each moment within the sliding analysis window. Integrate them according to the time sequence to obtain the time sequence working condition sequence. Model construction module: Based on the performance indicator vector at each moment, if the average attenuation analysis is triggered, an average performance attenuation model is constructed. If the dynamic working condition attenuation adaptation is triggered, the working condition cycle is combined with the time-series working condition sequence group within the stable attenuation period, and the working condition performance attenuation rate change model is constructed based on the time-series working condition sequence group; The average performance decay model is constructed by calculating the average decay rate of filtration efficiency and the average growth rate of resistance ratio during the stable decay period if the average decay analysis is triggered; The operating condition performance attenuation rate change model is formed by fitting the attenuation rate change model in different operating condition cycles according to the operating period if the dynamic operating condition attenuation adaptation is triggered; Prediction and deduction module: Use the obtained average performance attenuation model or operating performance attenuation rate change model to predict the remaining life of the vehicle air-conditioning filter element.

2. The system for predicting the life of a vehicle air conditioning filter according to claim 1, characterized in that: The method for judging whether the vehicle air conditioning filter is in the stable attenuation period is as follows: Obtain the result of determining whether the air conditioning filter is in a stable period. If the air conditioning filter is not in a stable period, calculate the Euclidean distance between the performance index vector of the air conditioning filter at the current moment and the previous moment, marking it as the adjacent attenuation distance. Calculate the Euclidean distance between the performance index vector of the air conditioning filter and the preset performance index vector bottom line, marking it as the attenuation end distance. If the adjacent attenuation distance at the current moment is less than or equal to the preset instantaneous change standard, and the attenuation end point distance at the current moment is greater than or equal to the preset distance standard, it is determined that the vehicle air-conditioning filter is in a stable attenuation period.

3. The vehicle air conditioning filter life prediction system according to claim 2, characterized in that: The method for judging whether the vehicle air conditioning filter is in the stable period is as follows: Obtain the performance index vector of the vehicle air conditioning filter at the starting point of its use, mark it as the initial performance index vector, and calculate the Euclidean distance between the performance index vector of the vehicle air conditioning filter and the initial performance index vector. If the Euclidean distance is less than a preset distance standard, determine that the vehicle air conditioning filter is in a stable period.

4. The vehicle air conditioning filter life prediction system according to claim 1, characterized in that: The method for determining whether to trigger dynamic working condition attenuation adaptation is as follows: The autocorrelation function is used to analyze the regularity of the time series operating condition sequence to analyze whether the time series operating condition sequence has periodicity. If the time series operating condition sequence has significant periodicity, the operating condition cycle is obtained according to the autocorrelation function. If there are at least two complete operating condition cycles in the stable attenuation period, it is determined that the dynamic operating condition attenuation adaptation is triggered.

5. The vehicle air conditioning filter life prediction system according to claim 1, characterized in that: The average performance decay model is constructed as follows: The performance index vector at the starting point of the stable attenuation period and the performance index vector at the current moment are obtained respectively. The performance index vector includes filtration efficiency and resistance ratio. The average attenuation rate of filtration efficiency and the average growth rate of resistance ratio during the stable attenuation period are calculated. Combined with the performance index vector at the current moment, a model of the change of filtration efficiency and resistance ratio over time is constructed, and the average performance attenuation model is obtained by integration.

6. The vehicle air conditioning filter life prediction system according to claim 1, characterized in that: The method for obtaining the timing period working condition sequence group is as follows: Consecutive moments of the same operating condition type are combined into one operating condition period. The same operating condition type is the operating condition type of the operating condition period. The performance index vectors of the starting and ending points of each operating condition period within the stable attenuation period are obtained. The average attenuation rate of filtration efficiency and the average growth rate of resistance ratio within the operating condition period are calculated. Obtain all working condition periods within a working condition cycle and integrate them according to the time sequence and mark the sequence numbers in sequence to obtain a time period working condition sequence. All time period working condition sequences within the stable attenuation period are integrated according to the time sequence and marked with sequence numbers in sequence to obtain a time period working condition sequence group. Both the cycle sequence number and the sequence number are integer counted starting from 1.

7. The vehicle air conditioning filter life prediction system according to claim 6, characterized in that: The construction method of the working condition performance attenuation rate change model is as follows: In the time-series cycle operating condition sequence group, all operating condition periods with the same sequence number in the cycle are integrated according to the time sequence to obtain the time-series operating condition period sequence with the sequence number in the corresponding cycle; In each time-series operating condition period sequence, the sequence number corresponding to the operating condition period is taken as the independent variable, and the average attenuation rate of filtration efficiency and the average growth rate of resistance ratio in the operating condition period are taken as dependent variables. The linear relationship is fitted by the least squares method to obtain the filtration efficiency attenuation rate change model and the resistance ratio growth rate change model of the sequence number in each cycle, and the results are summarized and integrated to obtain the operating condition performance attenuation rate change model.

8. The system for predicting the life of a vehicle air conditioning filter according to claim 7, characterized in that: The remaining life of the vehicle air conditioning filter element is predicted using the operating performance attenuation rate change model as follows: Set a deduction period with the current moment as the starting point. If the current moment is not the end of the operating cycle, predict and complete the current operating cycle within the deduction period. Use the operating performance attenuation rate change model to sequentially calculate the Euclidean distance between the performance index vector at the end of each operating period within the deduction period and the preset performance index vector baseline, marking it as the predicted life end distance. If the predicted life end distance is less than the preset distance standard, the deduction stops. If the predicted life end distance at the end of the current operating cycle is still greater than or equal to the preset distance standard, the next operating cycle will continue to be deduced within the deduction period until a working period is found that stops the deduction. This is marked as the end operating period. The predicted life end distances of the moments within the end operating period are calculated in sequence according to the time sequence. If the predicted life end distance at any moment is less than the preset distance standard, the moment is determined to be the end of the life of the vehicle air conditioning filter element, and subsequent moments are no longer calculated. Calculate the time interval between the current moment and the end of the lifespan to obtain the remaining lifespan of the vehicle air conditioning filter.

Images