CN-122020334-A - Clutch friction element deformation fault identification method, electronic equipment and storage medium

CN122020334ACN 122020334 ACN122020334 ACN 122020334ACN-122020334-A

Abstract

The invention provides a clutch friction element deformation fault identification method, electronic equipment and a storage medium. The method comprises the steps of obtaining an operation vibration response sequence of a target friction element under a preset working condition, converting the operation vibration response sequence into a frequency domain signal to obtain an original response spectrum, determining self-adaptive filtering weights of all frequency bands in the original response spectrum based on a pre-built frequency band discrimination evaluation model, carrying out weighted correction on the original response spectrum by utilizing the self-adaptive filtering weights to obtain a characteristic enhancement spectrum, calculating a spectrum energy disorder index based on the energy distribution probability of the characteristic enhancement spectrum, inputting the spectrum energy disorder index into a fault recognition classifier, and judging the deformation fault type of the target friction element. Therefore, the deformation fault recognition efficiency of the clutch friction element is improved.

Inventors

XUE JIAQI
SHI LUBING
LIU XIAOKUN
DING HAOYUAN
LIN RUIYI
JIANG HEXIN
XING HECHEN

Assignees

郑机所(郑州)传动科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260122

Claims (10)

1. A method for identifying deformation faults of a friction element of a clutch, comprising: Acquiring an operation vibration response sequence of a target friction element under a preset working condition; Converting the operation vibration response sequence into a frequency domain signal to obtain an original response frequency spectrum; Determining the self-adaptive filtering weight of each frequency band in the original response frequency spectrum based on a pre-constructed frequency band distinguishing degree evaluation model, wherein the construction logic of the frequency band distinguishing degree evaluation model is that the frequency spectrum characteristics of each reference sample in different deformation states are analyzed, if a certain frequency band shows high homogeneity characteristics in different deformation states, the frequency band is judged to be a noise leading frequency band, and the corresponding self-adaptive filtering weight is determined to be a suppression coefficient, and if a certain frequency band shows low homogeneity characteristics in different deformation states, the frequency band is judged to be a characteristic sensitive frequency band, and the corresponding self-adaptive filtering weight is determined to be a retention coefficient; Performing weighted correction on the original response spectrum by utilizing the self-adaptive filtering weight to obtain a characteristic enhancement spectrum; calculating a spectrum energy unordered index based on the energy distribution probability of the characteristic enhanced spectrum; and inputting the spectral energy disorder index into a fault identification classifier, and judging the deformation fault type of the target friction element.
2. The method for identifying deformation faults of clutch friction elements according to claim 1, wherein the analyzing spectral characteristics of each reference sample in different deformation states comprises: calculating the spectrum vector distance and the spectrum vector deflection angle value of any two groups of reference samples with different deformation states on the same frequency band; Synthesizing to obtain a difference score between states of the frequency band based on the spectrum vector distance and the spectrum vector deflection angle value; if the difference score between the states is smaller than a preset homogeneity judging threshold, the signal characteristics under the frequency band tend to be consistent between the two groups of states, and the frequency band is judged to show the high homogeneity characteristics; and if the inter-state difference score is greater than or equal to the homogeneity determination threshold, indicating that the signal characteristic under the frequency band has a significant distinction, and determining that the frequency band shows the low homogeneity characteristic.
3. The method of claim 2, wherein the band discrimination evaluation model further comprises global differential weighting logic: For the case that K different deformation states exist, respectively calculating the inter-state difference scores of the mth frequency band under all possible state combinations; Carrying out weighted summation on all the inter-state difference scores to obtain a global differential total score of the mth frequency band; if the global degree of distinction total score is located in the low-level interval of the score sequences of all the frequency bands, confirming that the frequency bands cannot effectively represent the deformation difference, and determining the corresponding self-adaptive filtering weight of the frequency bands to be zero or a value approaching to zero so as to eliminate the influence of the frequency bands in the subsequent steps.
4. A method of identifying a deformation failure of a clutch friction element according to claim 3, wherein said weighting the original response spectrum with the adaptive filter weights comprises: Constructing a frequency domain mask vector which corresponds to the original response frequency spectrum frequency points one by one, wherein element values in the frequency domain mask vector are the self-adaptive filtering weights; performing point-by-point multiplication operation on the original response spectrum and the frequency domain mask vector; If the adaptive filtering weight corresponding to a certain frequency point is the suppression coefficient, the amplitude of the frequency point is attenuated, so that the background noise interference of the frequency point is suppressed in the generated characteristic enhancement frequency spectrum; If the adaptive filtering weight corresponding to a certain frequency point is the retention coefficient, the amplitude of the frequency point is maintained or amplified, so that the vibration characteristic related to the deformation fault is highlighted.
5. The method of claim 1, wherein the calculating a spectral power disorder index comprises: Normalizing the characteristic enhancement spectrum to obtain the relative probability duty ratio of the amplitude of each frequency point in the total energy; Calculating a complexity measure of the feature enhanced spectrum using an information entropy algorithm based on the relative probability duty cycle; if the calculated complexity measurement value is larger than a preset measurement threshold value, the energy distribution state of the characteristic enhancement frequency spectrum is judged to be chaotic, and the higher the corresponding spectrum energy disorder index is, the more obvious nonlinear influence of the deformation degree of the characterization friction element on the system dynamics characteristic is.
6. The method for identifying deformation faults of a clutch friction element according to claim 1, further comprising establishing a reference sample deformation level, in particular: uniformly selecting a plurality of measuring points along the circumferential direction at the inner edge of the preselected reference friction element, and measuring the axial warping height of each measuring point relative to the outer edge datum plane; calculating the average value of the axial warping heights of all the measuring points to be used as a deformation quantization index; If the deformation quantification index is located in a first preset interval, judging that the friction element is a health reference sample; If the deformation quantization index is positioned in the second preset interval, judging that the friction element is a micro-deformation reference sample; If the deformation quantization index is positioned in a third preset interval, judging that the friction element is a serious deformation reference sample; The health reference sample, the micro-deformation reference sample and the severe deformation reference sample together form a training data set for constructing the frequency band discrimination evaluation model.
7. The method for identifying deformation faults of a clutch friction element according to claim 1, wherein the step of obtaining an operation vibration response sequence of a target friction element under a preset working condition comprises the following steps: Building a clutch separation working condition test bench, and installing a target friction element in a clutch box; when a target friction element is installed in a clutch box, considering constraint condition differences of different positions in the clutch, judging that the position, close to one side of a clamp spring, in the clutch box is a deformation sensitive position; controlling a brake to lock the steel sheet, and driving an input shaft to drive the friction plate to rotate so as to generate a relative rotation speed difference between the friction plate and the steel sheet; Triggering a vibration acquisition system when the rotating speed of the input shaft is stabilized at a rotating speed value corresponding to the preset working condition; And synchronously acquiring vibration signals of the clutch shell by using orthogonally arranged sensors, judging whether the acquisition time length reaches the preset sample window length, and if so, stopping acquisition and outputting the operation vibration response sequence.
8. The method of claim 1, wherein said inputting the spectral power disorder index into a fault identification classifier, determining the deformation fault class of the target friction element comprises: Taking the spectral energy disorder index of the reference sample of the known deformation fault class as a training set to be input into a classification model; Optimizing decision boundaries of the classification model by using the training set, so that entropy value intervals corresponding to different deformation degrees can be distinguished; Mapping the spectral energy disorder index of a target friction element into the decision boundary; judging the attribution of the region in which the spectrum energy disorder index falls, if the region falls into a first region, outputting a health state identification result, and if the region falls into a second region, outputting a deformation fault and a grade label corresponding to the deformation fault.
9. An electronic device, comprising: At least one processor; a memory communicatively coupled to the at least one processor; Wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.
10. A non-transitory computer readable storage medium storing computer instructions for causing the computer to perform the method of any one of claims 1-8.

Description

Clutch friction element deformation fault identification method, electronic equipment and storage medium Technical Field The present invention relates to the field of clutch friction element deformation fault recognition technologies, and in particular, to a method for recognizing a deformation fault of a clutch friction element, an electronic device, and a storage medium. Background The clutch friction element deformation fault is identified, the clutch friction element deformation is a key link for guaranteeing stable operation of a transmission system, the friction element deformation can directly lead to unsmooth engagement, slipping or incomplete separation of the clutch, power transmission loss and gear shifting blocking are caused, even abrasion of related parts such as a flywheel and a pressure plate is aggravated, power interruption in running of a vehicle can be caused when the friction element deformation fault is severe, potential safety hazards are buried, the fault is timely identified, mechanical linkage damage can be avoided in advance, maintenance cost is reduced, meanwhile, transmission efficiency and operation reliability of the clutch are guaranteed, driving safety is prevented from being influenced due to sudden failure, and the clutch friction element deformation has important practical significance for maintaining overall performance of the vehicle and prolonging service life of the transmission system. At present, the prior art is used for identifying the deformation faults of the friction element of the clutch, is mostly dependent on manual detection or single sensor data, is difficult to capture dynamic deformation characteristics in real time, is easy to vibrate and interfere by signals, has low characteristic extraction precision, and has low identification efficiency. Therefore, there is a need for a clutch friction element deformation failure recognition method with high recognition efficiency. Disclosure of Invention In order to solve the problems in the prior art, the invention provides a clutch friction element deformation fault identification method, electronic equipment and a storage medium. According to a first aspect of the present invention, a clutch friction element deformation failure recognition method is provided. The method comprises the following steps: Acquiring an operation vibration response sequence of a target friction element under a preset working condition; Converting the operation vibration response sequence into a frequency domain signal to obtain an original response frequency spectrum; Determining the self-adaptive filtering weight of each frequency band in the original response frequency spectrum based on a pre-constructed frequency band distinguishing degree evaluation model, wherein the construction logic of the frequency band distinguishing degree evaluation model is that the frequency spectrum characteristics of each reference sample in different deformation states are analyzed, if a certain frequency band shows high homogeneity characteristics in different deformation states, the frequency band is judged to be a noise leading frequency band, and the corresponding self-adaptive filtering weight is determined to be a suppression coefficient, and if a certain frequency band shows low homogeneity characteristics in different deformation states, the frequency band is judged to be a characteristic sensitive frequency band, and the corresponding self-adaptive filtering weight is determined to be a retention coefficient; Performing weighted correction on the original response spectrum by utilizing the self-adaptive filtering weight to obtain a characteristic enhancement spectrum; calculating a spectrum energy unordered index based on the energy distribution probability of the characteristic enhanced spectrum; and inputting the spectral energy disorder index into a fault identification classifier, and judging the deformation fault type of the target friction element. Further, the analyzing the spectral features of each reference sample under different deformation states includes: calculating the spectrum vector distance and the spectrum vector deflection angle value of any two groups of reference samples with different deformation states on the same frequency band; Synthesizing to obtain a difference score between states of the frequency band based on the spectrum vector distance and the spectrum vector deflection angle value; if the difference score between the states is smaller than a preset homogeneity judging threshold, the signal characteristics under the frequency band tend to be consistent between the two groups of states, and the frequency band is judged to show the high homogeneity characteristics; and if the inter-state difference score is greater than or equal to the homogeneity determination threshold, indicating that the signal characteristic under the frequency band has a significant distinction, and determining that the frequency band shows