A method for monitoring working condition of automobile radiator fins

By calculating the abnormal factors and filtering degree of vibration data and noise data of automobile heat dissipation fins, adaptive denoising processing solves the monitoring accuracy problem under the influence of noise data in the existing technology, and realizes the working status of automobile heat dissipation fins Accurate monitoring and alarm reminder ensure the stability of the heat dissipation function.

CN118500538BActive Publication Date: 2024-11-01CHIPING LUHUAN AUTOMOBILE RADIATOR CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410961743.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-18
Publication Date
2024-11-01
Estimated Expiration
2044-07-18

AI Technical Summary

Technical Problem

When the prior art uses temperature data and wind speed data to monitor the operating status of the car radiator, the influence of noise data is not effectively considered, resulting in low accuracy of the monitoring results.

Method used

By obtaining the vibration data and noise data of the car cooling fins when working, calculating the abnormality factor and filtering degree of the vibration data, adaptively adjusting the number of denoising processing times, using the Pearson correlation coefficient to eliminate the influence of the noise data, and then performing abnormal detection for monitoring Working status.

Benefits of technology

Accurate monitoring of the working conditions of the car's cooling fins is achieved, excessive or undersmoothing is avoided in the denoising process, the accuracy of the monitoring results is improved, and alarms are issued in a timely manner to ensure the stability of the heat dissipation function.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118500538B_ABST
    Figure CN118500538B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of data processing, and more specifically, to a method for monitoring the working condition of automobile radiating fins, the method comprising: obtaining a vibration data sequence and a noise data sequence, calculating an abnormal factor of any vibration data in the vibration data sequence, and calculating the filtering degree of each vibration data by comprehensively referring to a data segment, and adaptively adjusting the number of times each vibration data is subjected to denoising according to a comparison result between the filtering degree of each vibration data and a preset threshold, and performing anomaly detection on the vibration data sequence after denoising, so as to monitor the working condition of automobile radiating fins based on the result of the anomaly detection. The present invention can adaptively adjust the number of times each vibration data is subjected to denoising, can avoid over-smoothing and under-smoothing when denoising the vibration data, reduces the influence of the noise data, and can accurately monitor the working condition of automobile radiating fins.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data processing, and more specifically, to a method for monitoring the working condition of a car radiating fin. Background Art

[0002] Modern automobiles need to maintain high reliability and long-term operation under various complex working conditions. The stability of the cooling system is the basis for ensuring the normal operation of the engine and other key components. Abnormal vibrations when the automobile cooling fins are working may cause poor contact between the cooling fins and the radiator body, thereby affecting the heat conduction efficiency and reducing the heat dissipation effect to a certain extent. Therefore, by analyzing the abnormal conditions of the vibration data collected when the automobile cooling fins are working, the working condition of the cooling fins can be monitored, so that problems can be discovered and repaired in time.

[0003] In the related art, for example, a patent application document with publication number CN117632665A discloses a real-time monitoring method for the operation status of a radiator. The method comprises: collecting temperature data and wind speed data of the radiator, determining the possibility of abnormal change according to the correlation between the temperature data and the wind speed data, and the corresponding temperature abnormality degree, and determining the possibility of abnormal change according to each sampling moment. The abnormal degree of the temperature at the sampling time corresponding to the temperature of the neighboring radiator and the possibility of abnormal change are used to obtain the abnormal degree of the radiator, so as to monitor the operation status of the radiator using the abnormal degree of the radiator.

[0004] However, when the above scheme uses temperature data and wind speed data to monitor the operating status of the radiator, it does not take into account the influence of noise data in the acquired temperature data and wind speed data. The accuracy of the abnormality degree of the radiator obtained may be low, which may reduce the accuracy of monitoring the operating status of the radiator to a certain extent. Summary of the invention

[0005] In order to solve the problem of low accuracy of monitoring results due to the influence of noise data, the present invention provides a method for monitoring the working condition of automobile radiating fins.

[0006] According to one aspect of an embodiment of the present invention, a method for monitoring the working condition of a vehicle heat sink fin is provided, comprising:

[0007] Obtain the vibration data sequence and noise data sequence of the automobile radiator fin when it is working;

[0008] Calculate the abnormal factor of any vibration data in the vibration data sequence, the abnormal factor reflects the possibility that the vibration data is abnormal data;

[0009] The data segment in the noise data sequence that corresponds to the preset vibration data segment is used as the reference data segment to calculate the filtering degree of each vibration data. , For the The abnormal factors of vibration data, To preset vibration data segment Neidi The percentile of the vibration data, For reference data segment Neidi The percentile of the noise data collected at the corresponding moment of each vibration data, For the preset vibration data segment except The data other than the vibration data are the same as the reference data segment except the The noise data collected at the corresponding moment of each vibration data The Pearson correlation coefficient between the data outside is the Pearson correlation coefficient between the data in the preset vibration data segment and the data in the reference data segment, is the normalization function, is the absolute value symbol;

[0010] According to the comparison result between the filtering degree of each vibration data and the preset threshold, the number of times the vibration data is denoised is adaptively adjusted, and the vibration data sequence after denoising is performed anomaly detection to monitor the working condition of the automobile cooling fin based on the results of anomaly detection.

[0011] The present invention adaptively adjusts the number of times denoising is performed on each vibration data by comparing the filtering degree of the vibration data with a preset threshold, thereby avoiding over-smoothing and under-smoothing when denoising the vibration data, thereby accurately denoising the vibration data sequence, and further accurately monitoring the working condition of the automobile radiator fin.

[0012] In an exemplary embodiment of the present invention, according to the comparison result of the filtering degree of each vibration data and the preset threshold, the number of times of performing denoising processing on each vibration data is adaptively adjusted, including:

[0013] When the filtering degree of the vibration data is greater than or equal to the preset threshold, the vibration data is denoised using the mean filtering algorithm until the filtering degree of the vibration data after the current denoising process is less than the preset threshold, so as to adaptively adjust the number of denoising processes performed on each vibration data.

[0014] In an exemplary embodiment of the present invention, anomaly detection is performed on the vibration data sequence after denoising to monitor the working condition of the automobile radiator fin based on the result of the anomaly detection, including:

[0015] Determine the normal range of vibration data in the vibration data sequence after denoising. When the value of any vibration data is outside the normal range, the vibration data is judged to be abnormal vibration data, and an alarm is issued to remind the operator that the automobile cooling fin has an abnormality at the corresponding moment, thereby monitoring the working condition of the automobile cooling fin.

[0016] The present invention can accurately monitor the working condition of the automobile radiating fins, and send out an alarm message when an abnormal situation is detected, so that the operator can take corresponding measures to repair the automobile radiating fins in time, thereby ensuring the stability of the heat dissipation function.

[0017] In an exemplary embodiment of the present invention, a method for obtaining a normal range includes:

[0018] A box plot is constructed based on the vibration data sequence after denoising, and the range consisting of the upper limit and the lower limit of the box plot is used as the normal range of the vibration data in the vibration data sequence after denoising.

[0019] The present invention uses a box plot to determine the normal range of vibration data in a vibration data sequence, thereby ensuring the accuracy of the determined normal range.

[0020] In an exemplary embodiment of the present invention, a method for obtaining an abnormal factor of any vibration data in a vibration data sequence includes:

[0021] The absolute value of the difference between the median of the data in the preset vibration data segment and the average of the data in the preset vibration data segment is used as the deviation degree of the data in the preset vibration data segment;

[0022] Randomly select multiple vibration data segments with the same length as the preset vibration data segment from the vibration data sequence, calculate the cumulative sum of the differences between the information entropy of the preset vibration data segment and the information entropy of each vibration data segment, and use the cumulative sum as the credibility of the deviation degree;

[0023] The abnormal factor of any vibration data is calculated, and the abnormal factor is positively correlated with the degree of deviation, the credibility of the degree of deviation, and the abnormality of the value of any vibration data.

[0024] The present invention determines the abnormal factors of vibration data by integrating multiple aspects, thereby ensuring the accuracy of the abnormal factors.

[0025] In an exemplary embodiment of the present invention, the abnormal factor of the vibration data satisfies the following relationship:

[0026] ;

[0027] In the formula, For the Abnormal factors of vibration data; For the The value of vibration data; It is the average value of the data in the preset vibration data segment; It is the median of the data in the preset vibration data segment; is the information entropy of the data in the preset vibration data segment; The same length as the preset vibration data segment The information entropy of the data in a vibration data segment; is the normalization function; is the absolute value symbol.

[0028] In an exemplary embodiment of the present invention, a method for obtaining percentiles includes:

[0029] Arrange the data in a preset vibration data segment of any vibration data in order or in reverse order to obtain a position sequence number of any vibration data;

[0030] Calculate the ratio of the position number to the end position number corresponding to the preset vibration data segment, and divide the ratio into percentages to obtain the percentile of any vibration data in the preset vibration data segment. Similarly, obtain the percentile of the noise data collected at the corresponding moment of any vibration data in the reference data segment.

[0031] The present invention uses the concept of percentile in statistics to measure the correlation between vibration data and noise data at the same time, which can eliminate the influence of different dimensions and different orders of magnitude and ensure the accuracy of the determined correlation.

[0032] In an exemplary embodiment of the present invention, a method for obtaining a preset vibration data segment includes:

[0033] Taking any vibration data in the vibration data sequence as the center, a vibration data segment with a preset length is selected to obtain a preset vibration data segment of any vibration data, and then obtain a preset vibration data segment corresponding to each vibration data.

[0034] The present invention has the following effects:

[0035] The present invention combines various data such as abnormal factors of vibration data and correlation between vibration data and noise data collected at the same time to calculate the filtering degree of vibration data, thereby ensuring the accuracy of the filtering degree. The present invention also adjusts the number of times denoising is performed on each vibration data according to the filtering degree of each vibration data, thereby avoiding over-smoothing and under-smoothing when denoising is performed on each vibration data, reducing the influence of noise data, and accurately monitoring the working condition of the automobile radiating fin. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0037] Figure 1 It is a schematic flow chart of the steps of a method for monitoring the working condition of a heat sink fin of an automobile according to an embodiment of the present invention;

[0038] Figure 2 It is a flowchart of the steps of determining abnormal factors of vibration data in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0040] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0041] Reference Figure 1 , a method for monitoring the working condition of a car radiating fin, comprising steps S1 to S4, specifically as follows:

[0042] S1: Obtain the vibration data sequence and noise data sequence of the automobile cooling fin when it is working.

[0043] In an exemplary embodiment of the present invention, the vibration data sequence may be a data sequence consisting of readings from a vibration sensor electrically connected to the automobile radiating fin, used as a data basis for monitoring the working condition of the automobile radiating fin. The noise data sequence may be a data sequence consisting of readings from a sound level meter electrically connected to the automobile radiating fin, used as control data for further analyzing the vibration data.

[0044] Optionally, the readings from the vibration sensor electrically connected to the automobile cooling fins and the sound level meter can be continuously collected at a fixed frequency such as 1 time / second within a certain time period, such as 30 minutes, to obtain vibration data and noise data within the time period, and the data sequence composed of all the vibration data within the time period can be used as a vibration data sequence, and the data sequence composed of all the noise data within the time period can be used as a noise data sequence; this embodiment does not specifically limit the collection time and collection frequency.

[0045] S2: Calculate the abnormal factor of any vibration data in the vibration data sequence. The abnormal factor reflects the possibility that the vibration data is abnormal data.

[0046] In an exemplary embodiment of the present invention, the abnormal factor refers to data used to measure the possibility that any vibration data is abnormal data. When the abnormal factor of any vibration data is larger, the possibility that the vibration data is abnormal data is greater. On the contrary, when the abnormal factor of any vibration data is smaller, the possibility that the vibration data is abnormal data is smaller.

[0047] In an exemplary embodiment of the present invention, reference Figure 2 As shown, the abnormal factor of any vibration data in the vibration data sequence can be determined by the following steps:

[0048] S101: taking the absolute value of the difference between the median of the data in the preset vibration data segment and the average value of the data in the preset vibration data segment as the deviation degree of the data in the preset vibration data segment;

[0049] Among them, the preset vibration data segment can be a vibration data segment determined with any vibration data as the center, which is used as a data basis to analyze the possibility that the corresponding vibration data is abnormal data. The degree of deviation can be data that measures the distribution of data in the preset vibration data segment, which is used to determine the possibility that abnormal data exists in the preset vibration data segment. If the degree of deviation of the preset vibration data segment determined with any vibration data as the center is large, it means that the data distribution in the preset vibration data segment is relatively discrete, which further means that the possibility of abnormal data existing in the preset vibration data segment is large, and the possibility that the vibration data is abnormal data is relatively large; on the contrary, the possibility that the vibration data is abnormal data is relatively small.

[0050] In an exemplary embodiment of the present invention, the determination of the preset vibration data segment can be achieved by the following steps:

[0051] Taking any vibration data in the vibration data sequence as the center, a vibration data segment with a preset length is selected to obtain a preset vibration data segment of any vibration data, and then obtain a preset vibration data segment corresponding to each vibration data.

[0052] Exemplarily, the preset length can be set to 101. When determining the preset vibration data segment of any vibration data, a vibration data segment with a length of 101 selected with the vibration data as the center will be used as the preset vibration data segment of the vibration data. Of course, a suitable length can also be set according to the specific situation. This embodiment does not specifically limit the length of the preset vibration data segment.

[0053] Optionally, when the number of vibration data on either side of the vibration data is 49 and less than 50, a data segment consisting of 49 vibration data on that side, the vibration data and 51 vibration data on the other side is used as a preset vibration data segment, thereby determining the preset vibration data segment corresponding to each vibration data in the vibration data sequence.

[0054] S102: randomly selecting a plurality of vibration data segments having the same length as a preset vibration data segment from the vibration data sequence, calculating the cumulative sum of the differences between the information entropy of the data in the preset vibration data segment and the information entropy of the data in each vibration data segment, and using the cumulative sum as the credibility of the offset degree;

[0055] The credibility of the degree of deviation refers to the cumulative sum of the difference between the information entropy of the data in the preset vibration data segment and the information entropy of the data in each vibration data segment, and is used to measure the accuracy of the degree of deviation of the preset vibration data segment. For example, when it is determined that the degree of deviation of the preset vibration data segment of any vibration data is large, if the credibility of the degree of deviation is also large, the accuracy of the degree of deviation is high, indicating that there is a high possibility of abnormal data in the preset vibration data segment of the vibration data; if the credibility of the degree of deviation is small, the accuracy of the degree of deviation is low, indicating that there is a relatively small possibility of abnormal data in the preset vibration data segment of the vibration data, thereby ensuring the accuracy of the possibility of abnormal data in the preset vibration data segment.

[0056] S103: Calculate an abnormal factor of any vibration data. The abnormal factor is positively correlated with the degree of deviation, the credibility of the degree of deviation, and the abnormality of the value of any vibration data.

[0057] Specifically, the abnormal factor of vibration data satisfies the following relationship:

[0058] ;

[0059] In the formula, For the Abnormal factors of vibration data; For the The value of vibration data; It is the average value of the data in the preset vibration data segment; It is the median of the data in the preset vibration data segment; is the information entropy of the preset vibration data segment; The same length as the preset vibration data segment The information entropy of each vibration data segment; is the normalization function; is the absolute value symbol.

[0060] in, It reflects the abnormality of the vibration data value. The larger the value is, the greater the abnormality of the vibration data value is, indicating that the vibration data is more likely to be abnormal data, and the corresponding abnormal factor of the vibration data is larger; It reflects the degree of deviation of the preset vibration data segment of the vibration data. The larger the value is, the greater the possibility that abnormal data exists in the preset vibration data segment of the vibration data, and the larger the abnormal factor of the corresponding vibration data is. It reflects the credibility of the deviation degree of the preset vibration data segment of the vibration data. The larger the value is, the greater the possibility that abnormal data exists in the preset vibration data segment of the vibration data, and the larger the corresponding abnormal factor of the vibration data.

[0061] In another embodiment, the abnormal factor of each vibration data in the vibration data sequence may be determined by other formulas, such as by a summation or gamma transformation formula instead of a product calculation formula.

[0062] S3: taking the data segment in the noise data sequence that has the same time period as the preset vibration data segment as the reference data segment, and calculating the filtering degree of each vibration data.

[0063] It should be noted that when collecting vibration data when the automobile radiating fins are working, the electronic equipment inside the vehicle and the external electromagnetic radiation source may generate electromagnetic interference to the vibration sensor, resulting in noise data in the collected vibration data. The abnormal factor of each vibration data reflects the possibility that the corresponding vibration data is abnormal data, and it is impossible to distinguish between noise data and abnormal vibration data caused by abnormalities in the automobile radiating fins. Therefore, it is necessary to determine the possibility that each vibration data is noise data, so as to accurately denoise the vibration data sequence.

[0064] In an exemplary embodiment of the present invention, the filtering degree refers to data that can reflect the possibility that each vibration data is noise data, and is used as a basis for judging whether to perform denoising on each vibration data. For example, when the filtering degree of any vibration data is large, it means that the vibration data is likely to be noise data, and when the filtering degree of the vibration data is greater than a certain value, the vibration data is denoised, thereby reducing the impact of the noise data on monitoring the working condition of the automobile radiator fin.

[0065] Specifically, the filtering degree of each vibration data satisfies the following relationship:

[0066] ;

[0067] In the formula, For the The degree of filtering of vibration data; For the Abnormal factors of vibration data; To preset vibration data segment Neidi Percentile of vibration data; For reference data segment Neidi The percentile of the noise data collected at the corresponding moment of each vibration data; For the preset vibration data segment except The data other than the vibration data are the same as the reference data segment except the The noise data collected at the corresponding moment of each vibration data Pearson correlation coefficient between the data outside is the Pearson correlation coefficient between the data in the preset vibration data segment and the data in the reference data segment. It should be noted that the determination of the Pearson correlation coefficient between the data is a prior art and will not be described in detail here; is the normalization function; is the absolute value symbol.

[0068] in, Reflects the The deviation of the relative positions of the vibration data and noise data collected at the corresponding moment of each vibration data in the preset vibration data segment and the reference data segment. The larger the value, the larger the corresponding deviation, and the lower the correlation between the vibration data and the noise data at this moment. If the vibration data is more likely to be abnormal data, then the vibration data is more likely to be noise data, and the corresponding filtering degree is greater; on the contrary, the smaller the value, the smaller the corresponding deviation, and the greater the correlation between the vibration data and the noise data at this moment. If the vibration data is more likely to be abnormal data, then the reason for the abnormality of the vibration data is more likely to be an abnormality of the automobile cooling fin, and the corresponding filtering degree is smaller.

[0069] Reflects the The larger the value is, the greater the corresponding influence is, indicating that the correlation between the vibration data and noise data collected at this moment is smaller. If the vibration data is more likely to be abnormal data, then the vibration data is more likely to be noise data, and the corresponding filtering degree is greater. Otherwise, the corresponding filtering degree is greater.

[0070] In an exemplary embodiment of the present invention, the determination of percentiles may be achieved by the following steps:

[0071] Arrange the data in a preset vibration data segment of any vibration data in ascending or descending order to obtain the position number of any vibration data; calculate the ratio of the position number to the end position number corresponding to the preset vibration data segment, and percentile the ratio to obtain the percentile of any vibration data in the preset vibration data segment. Similarly, obtain the percentile of the noise data collected at the corresponding moment of any vibration data in the reference data segment.

[0072] Optionally, by calculating the percentile of any vibration data in a preset vibration data segment and the percentile of the noise data collected at the corresponding moment of the vibration data in the reference data segment, the influence of different orders of magnitude and different dimensions of data when analyzing the correlation between vibration data and noise data can be eliminated, and the correlation between vibration data and noise data collected at the same moment can be accurately determined, thereby ensuring the accuracy of the filtering degree of each vibration data.

[0073] S4: According to the comparison result of the filtering degree of each vibration data with the preset threshold, the number of times of denoising for each vibration data is adaptively adjusted, and the vibration data sequence after denoising is performed anomaly detection, so as to monitor the working condition of the automobile cooling fin based on the result of anomaly detection.

[0074] In an exemplary embodiment of the present invention, the number of times of performing denoising processing on each vibration data can be adaptively adjusted by the following steps:

[0075] When the filtering degree of the vibration data is greater than or equal to the preset threshold, the vibration data is denoised using the mean filtering algorithm until the filtering degree of the vibration data after the current denoising process is less than the preset threshold, so as to adaptively adjust the number of denoising processes performed on each vibration data.

[0076] It should be noted that when using the mean filtering algorithm to denoise the vibration data sequence, if the denoising process is performed the same number of times on each vibration data in the vibration data sequence, some vibration data may be over-smoothed, resulting in loss of detail information, and some vibration data may be under-smoothed, resulting in a large denoising effect, which makes it impossible to accurately monitor the working condition of the automobile cooling fins. Therefore, it is necessary to adaptively adjust the number of denoising processes for each vibration data, so as to accurately denoise the vibration data sequence, and then accurately monitor the working condition of the automobile cooling fins.

[0077] Optionally, the preset threshold can be set to 0.6. When the filtering degree of any vibration data in the vibration data sequence is greater than or equal to 0.6, it indicates that the vibration data is more likely to be noise data, and it is necessary to use a mean filtering algorithm to denoise the vibration data. When the filtering degree of the vibration data after the current denoising process is less than 0.6, the vibration data will no longer be denoised. Thus, the number of denoising processes for each vibration data can be adaptively adjusted, and over-smoothing and under-smoothing phenomena can be avoided when denoising each vibration data, thereby accurately denoising the vibration data sequence. It should be noted that the mean filtering algorithm is a prior art and is not described in detail in this embodiment.

[0078] In an exemplary embodiment of the present invention, the working condition of the automobile radiating fin can be monitored by the following steps:

[0079] Determine the normal range of vibration data in the vibration data sequence after denoising. When the value of any vibration data is outside the normal range, the vibration data is judged to be abnormal vibration data, and an alarm is issued to remind the operator that the automobile cooling fin has an abnormality at the corresponding moment, thereby monitoring the working condition of the automobile cooling fin.

[0080] The normal range refers to the value range of normal vibration data in the vibration data sequence, which is used as a judgment condition for determining abnormal vibration data. Abnormal vibration data refers to the vibration data when an abnormality occurs in the automobile radiating fin, which is used as data for triggering alarm information to remind the operator that an abnormality occurs in the automobile radiating fin at the corresponding moment, so that the operator can take corresponding measures to repair the automobile radiating fin in time, thereby realizing the monitoring of the working condition of the automobile radiating fin.

[0081] In an exemplary embodiment of the present invention, the determination of the normal range of vibration data in the vibration data sequence may be achieved by the following steps:

[0082] A box plot is constructed based on the vibration data sequence after denoising, and the range consisting of the upper limit and the lower limit of the box plot is used as the normal range of the vibration data in the vibration data sequence after denoising.

[0083] Among them, when the corresponding box plot is constructed based on the vibration data sequence after denoising, if any data point in the box plot exceeds the upper limit or lower limit of the box plot, the data point is abnormal data. Therefore, the range composed of the upper limit and lower limit of the box plot is the normal range of the vibration data in the vibration data sequence after denoising. It should be noted that the method for constructing the box plot is a prior art, and the upper limit and lower limit of the box plot are professional terms in the box plot, which will not be described in detail in this implementation.

[0084] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0085] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for monitoring the working condition of automobile radiator fins, It is characterized in that include: Obtain the vibration data sequence and noise data sequence of the automobile radiator fin when it is working; Calculating an abnormal factor of any vibration data in the vibration data sequence, wherein the abnormal factor reflects the possibility that the vibration data is abnormal data; The data segment in the noise data sequence that has the same time period as the preset vibration data segment is used as the reference data segment, and the filtering degree of each vibration data is calculated. , For the The abnormal factors of vibration data, To preset vibration data segment Neidi The percentile of the vibration data, For reference data segment Neidi The percentile of the noise data collected at the corresponding moment of each vibration data, For the preset vibration data segment except The data other than the vibration data are the same as the reference data segment except the The noise data collected at the corresponding moment of each vibration data The Pearson correlation coefficient between the data outside is the Pearson correlation coefficient between the data in the preset vibration data segment and the data in the reference data segment, is the normalization function, is the absolute value symbol; According to the comparison result of the filtering degree of each vibration data and the preset threshold, the number of times of denoising the vibration data is adaptively adjusted, and the vibration data sequence after denoising is performed is abnormally detected, so as to monitor the working condition of the automobile radiating fin based on the result of abnormality detection; The method for obtaining the abnormal factor includes: The absolute value of the difference between the median of the data in the preset vibration data segment and the average of the data in the preset vibration data segment is used as the deviation degree of the data in the preset vibration data segment; Randomly selecting a plurality of vibration data segments having the same length as a preset vibration data segment from the vibration data sequence, calculating a cumulative sum of differences between information entropies of the preset vibration data segment and information entropies of each of the vibration data segments, and using the cumulative sum as the credibility of the offset degree; Calculating an abnormal factor of any vibration data, wherein the abnormal factor is positively correlated with the degree of deviation, the credibility of the degree of deviation, and an abnormality of a value of any vibration data; The abnormal factor satisfies the following relationship: , where For the Abnormal factors of vibration data; For the The value of vibration data; It is the average value of the data in the preset vibration data segment; It is the median of the data in the preset vibration data segment; is the information entropy of the data in the preset vibration data segment; The same length as the preset vibration data segment The information entropy of the data in each vibration data segment; is the normalization function; is the absolute value symbol; The adaptively adjusting the number of times the denoising process is performed on each of the vibration data comprises: when the filtering degree of the vibration data is greater than or equal to the preset threshold, the vibration data is denoised using a mean filtering algorithm until the filtering degree of the vibration data after the current denoising process is completed is less than the preset threshold, so as to adaptively adjust the number of times the denoising process is performed on each of the vibration data; The abnormality detection is performed on the vibration data sequence after denoising to monitor the working condition of the automobile radiating fin based on the result of the abnormality detection, including: determining the normal range of vibration data in the vibration data sequence after denoising, when the value of any vibration data is outside the normal range, the vibration data is determined to be abnormal vibration data, and an alarm is issued to remind the operator that the automobile radiating fin has an abnormality at the corresponding moment, thereby monitoring the working condition of the automobile radiating fin.

2. A method for monitoring the working condition of a vehicle radiating fin according to claim 1, It is characterized in that The method for obtaining the normal range includes: A box plot is constructed based on the vibration data sequence after denoising, and a range consisting of an upper limit and a lower limit of the box plot is used as a normal range of vibration data in the vibration data sequence after denoising.

3. A method for monitoring the working condition of automobile radiating fins according to claim 1, It is characterized in that The method for obtaining the percentile includes: Arrange the data in a preset vibration data segment of any vibration data in order or in reverse order to obtain a position sequence number of any vibration data; Calculate the ratio of the position number to the end position number corresponding to the preset vibration data segment, and percentile the ratio to obtain the percentile of any vibration data in the preset vibration data segment. Similarly, obtain the percentile of the noise data collected at the corresponding moment of any vibration data in the reference data segment.

4. A method for monitoring the working condition of automobile radiating fins according to claim 3, It is characterized in that The method for obtaining the preset vibration data segment includes: Taking any vibration data in the vibration data sequence as the center, a vibration data segment with a preset length is selected to obtain a preset vibration data segment of the any vibration data, and then obtain a preset vibration data segment corresponding to each vibration data.