Clutch friction element deformation fault identification method, electronic equipment and storage medium

Abstract

The invention provides a clutch friction element deformation fault identification method, electronic equipment and a storage medium. The method is applied to the technical field of clutch friction element deformation fault recognition and comprises the steps that an operation vibration response sequence of a target friction element under a preset working condition is obtained; converting the operation vibration response sequence into a frequency domain signal to obtain an original response frequency spectrum; based on a pre-constructed frequency band discrimination evaluation model, determining an adaptive filtering weight of each frequency band in the original response spectrum; performing weighted correction on the original response frequency spectrum by using the self-adaptive filtering weight to obtain a feature enhanced frequency spectrum; calculating a spectrum energy disorder degree index based on the energy distribution probability of the feature enhanced spectrum; and inputting the spectrum energy disorder degree index into a fault identification classifier, and judging the deformation fault category of the target friction element. Therefore, the deformation fault identification efficiency of the clutch friction element is improved.

Description

Technical Field

[0001] This invention relates to the field of clutch friction element deformation fault identification technology, and in particular to a clutch friction element deformation fault identification method, electronic device and storage medium. Background Technology

[0002] Identifying clutch friction element deformation faults is a crucial step in ensuring the stable operation of the transmission system. Deformation of friction elements can directly lead to uneven clutch engagement, slippage, or incomplete disengagement, causing power transmission losses, shifting difficulties, and even accelerating the wear of related components such as the flywheel and pressure plate. In severe cases, it can even cause power interruption while the vehicle is in motion, creating a safety hazard. Timely identification of this fault can prevent mechanical chain damage in advance, reduce maintenance costs, and at the same time ensure the transmission efficiency and operational reliability of the clutch. It also avoids the impact of sudden faults on driving safety, which is of great practical significance for maintaining the overall performance of the vehicle and extending the service life of the transmission system.

[0003] Currently, existing technologies for identifying deformation faults in clutch friction components mostly rely on manual inspection or data from a single sensor, making it difficult to capture dynamic deformation features in real time; the signals are easily affected by vibration and temperature interference, and the feature extraction accuracy is low, resulting in low identification efficiency.

[0004] Therefore, there is an urgent need for a method for identifying clutch friction element deformation faults with high identification efficiency. Summary of the Invention

[0005] To address the problems existing in the prior art, the present invention provides a method for identifying deformation faults in clutch friction elements, an electronic device, and a storage medium.

[0006] According to a first aspect of the present invention, a method for identifying deformation faults in clutch friction elements is provided. The method includes:

[0007] Obtain the vibration response sequence of the target friction element under preset operating conditions;

[0008] The vibration response sequence is converted into a frequency domain signal to obtain the original response spectrum;

[0009] Based on a pre-constructed frequency band discrimination evaluation model, the adaptive filtering weights of each frequency band in the original response spectrum are determined. The construction logic of the frequency band discrimination evaluation model is as follows: Analyze the spectral characteristics of each reference sample under different deformation states; if a frequency band exhibits high homogeneity across different deformation states, it is determined to be a noise-dominant frequency band, and its corresponding adaptive filtering weight is determined as a suppression coefficient; if a frequency band exhibits low homogeneity across different deformation states, it is determined to be a feature-sensitive frequency band, and its corresponding adaptive filtering weight is determined as a retention coefficient.

[0010] The original response spectrum is weighted and corrected using the adaptive filtering weights to obtain the feature-enhanced spectrum;

[0011] Based on the energy distribution probability of the enhanced spectrum, the spectral energy disorder index is calculated.

[0012] The spectral energy disorder index is input into the fault identification classifier to determine the deformation fault category of the target friction element.

[0013] Furthermore, the analysis of the spectral characteristics of each reference sample under different deformation states includes:

[0014] Calculate the spectral vector distance and spectral vector deflection angle between any two sets of reference samples in different deformation states in the same frequency band;

[0015] Based on the spectral vector distance and the spectral vector deflection angle, a state difference score for this frequency band is synthesized.

[0016] If the difference score between states is less than the preset homogeneity judgment threshold, it indicates that the signal characteristics of the frequency band tend to be consistent between the two sets of states, and the frequency band is judged to exhibit the high homogeneity characteristics.

[0017] If the difference score between states is greater than or equal to the homogenization determination threshold, it indicates that the signal features in this frequency band have significant distinguishability, and the frequency band is determined to exhibit the low homogenization feature.

[0018] Furthermore, the frequency band discrimination evaluation model also includes global difference weighting logic:

[0019] For cases with K different deformation states, calculate the state difference score for the m-th frequency band under all possible state combinations;

[0020] The total global discrimination score for the m-th frequency band is obtained by weighted summation of all state difference scores.

[0021] If the total global discrimination score is located in the low quantile range of all frequency band score sequences, it is confirmed that the frequency band cannot effectively represent the deformation difference, and its corresponding adaptive filtering weight is determined to be zero or close to zero, so as to remove the influence of the frequency band in subsequent steps.

[0022] Further, the weighted correction of the original response spectrum using the adaptive filtering weights includes:

[0023] Construct a frequency domain mask vector that corresponds one-to-one with the frequency points of the original response spectrum, and the element values ​​in the frequency domain mask vector are the adaptive filtering weights;

[0024] Perform point-by-point multiplication between the original response spectrum and the frequency domain mask vector;

[0025] If the adaptive filtering weight corresponding to a certain frequency point is the suppression coefficient, then the amplitude of that frequency point is attenuated, thereby suppressing the background noise interference of that frequency point in the generated feature enhancement spectrum;

[0026] If the adaptive filtering weight corresponding to a certain frequency point is the retention coefficient, then the amplitude of that frequency point is maintained or amplified, thereby highlighting the vibration characteristics related to deformation faults.

[0027] Furthermore, the calculation of the spectral energy disorder index includes:

[0028] The enhanced spectrum of the feature is normalized to obtain the relative probability of each frequency point amplitude in the total energy;

[0029] Based on the relative probability proportions, the complexity measure of the feature enhancement spectrum is calculated using the information entropy algorithm;

[0030] If the calculated complexity metric value is greater than the preset metric threshold, the energy distribution state of the feature-enhanced spectrum is determined to be chaotic. The higher the corresponding spectral disorder index, the more significant the nonlinear influence of the deformation degree of the friction element on the system dynamic characteristics.

[0031] Furthermore, before obtaining the vibration response sequence of the target friction element under preset operating conditions, the process also includes establishing a reference sample deformation level, specifically:

[0032] Several measurement points are uniformly selected along the circumferential direction on the inner edge of the pre-selected reference friction element, and the axial warping height of each measurement point relative to the outer edge reference surface is measured.

[0033] Calculate the average axial warping height of all measurement points as a deformation quantification index;

[0034] If the deformation quantification index is within the first preset range, the friction element is determined to be a healthy benchmark sample.

[0035] If the deformation quantification index is within the second preset range, then the friction element is determined to be a micro-deformation benchmark sample;

[0036] If the deformation quantification index is within the third preset range, the friction element is determined to be a severely deformed benchmark sample.

[0037] The healthy baseline samples, the slightly deformed baseline samples, and the severely deformed baseline samples together constitute the training dataset used to construct the frequency band discrimination evaluation model.

[0038] Furthermore, the acquisition of the operating vibration response sequence of the target friction element under preset operating conditions includes:

[0039] A test bench for clutch disengagement conditions was constructed, and the target friction element was installed inside the clutch housing. When installing the target friction element inside the clutch housing, considering the differences in constraint conditions at different locations inside the clutch, the position near the snap ring inside the clutch housing was determined to be a deformation-sensitive position. The target friction element was assembled at the deformation-sensitive position to ensure that the maximum amplitude of abnormal hydrodynamic vibration caused by warping deformation could be captured under disengagement conditions.

[0040] The control brake locks the steel plate and drives the input shaft to rotate the friction plate, creating a relative speed difference between the friction plate and the steel plate;

[0041] When the input shaft speed stabilizes at the speed value corresponding to the preset working condition, the vibration acquisition system is triggered;

[0042] Vibration signals of the clutch housing are synchronously acquired using orthogonally arranged sensors. The acquisition time is then determined to be within the preset sample window length. If the acquisition time is reached, the acquisition is stopped and the operating vibration response sequence is output.

[0043] Further, the step of inputting the spectral energy disorder index into the fault identification classifier to determine the deformation fault category of the target friction element includes:

[0044] The spectral energy disorder index of reference samples with known deformation fault categories is used as the training set input to the classification model.

[0045] The decision boundary of the classification model is optimized using the training set so that it can distinguish the entropy value range corresponding to different degrees of deformation.

[0046] The spectral energy disorder index of the target friction element is mapped to the decision boundary;

[0047] Determine the region to which the spectral disorder index falls. If it falls into the first region, output the health status identification result; if it falls into the second region, output the deformation fault and its corresponding level label.

[0048] According to a second aspect of the present invention, an electronic device is provided. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the program to implement the method.

[0049] According to a third aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method.

[0050] This invention distinguishes between noise-dominant frequency bands and feature-sensitive frequency bands through adaptive filtering weights, thereby enhancing the features and suppressing noise of the original response spectrum and improving the identification of deformable fault features. It also quantifies the energy distribution pattern of the enhanced spectrum by calculating the spectral energy disorder index, thereby achieving quantitative characterization of fault features and improving the accuracy of fault category determination. Finally, it classifies the quantified spectral energy disorder index using a fault identification classifier, thereby achieving automated determination of deformable fault categories and improving the overall efficiency of fault identification.

[0051] It should be understood that the description in the Summary of the Invention is not intended to limit the key or essential features of the embodiments of the present invention, nor is it intended to restrict the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0052] The above and other features, advantages, and aspects of the various embodiments of the present invention will become more apparent from the accompanying drawings and the following detailed description. The drawings are provided for a better understanding of the invention and are not intended to limit the scope of the invention. In the drawings, the same or similar reference numerals denote the same or similar elements, wherein:

[0053] Figure 1 A flowchart of a clutch friction element deformation fault identification method according to an embodiment of the present invention is shown;

[0054] Figure 2 A block diagram of an exemplary electronic device capable of implementing embodiments of the present invention is shown. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0057] Figure 1 A flowchart of a clutch friction element deformation fault identification method according to an embodiment of the present invention is shown. The method includes:

[0058] S101, Obtain the operating vibration response sequence of the target friction element under preset working conditions;

[0059] In some embodiments, before obtaining the vibration response sequence of the target friction element under preset working conditions, the method further includes establishing a deformation level of a reference sample. Specifically, this involves: uniformly selecting several measurement points along the circumferential direction on the inner edge of a pre-selected reference friction element, measuring the axial warping height of each measurement point relative to the outer edge reference surface; calculating the average value of the axial warping height of all measurement points as a deformation quantification index; if the deformation quantification index is within a first preset interval, the friction element is determined to be a healthy reference sample; if the deformation quantification index is within a second preset interval, the friction element is determined to be a slightly deformed reference sample; if the deformation quantification index is within a third preset interval, the friction element is determined to be a severely deformed reference sample; the healthy reference sample, the slightly deformed reference sample, and the severely deformed reference sample together constitute a training dataset for constructing the frequency band discrimination evaluation model. According to embodiments of the present invention, by establishing a standardized deformation quantification index based on the average axial warping height, the random error of a single measurement point is eliminated, achieving accurate and quantitative characterization of the degree of deformation. A structured, hierarchical training dataset is constructed by dividing the data into three levels of reference samples: healthy, slightly deformed, and severely deformed. This provides clear classification labels and benchmark features for the frequency band discrimination evaluation model. When the vibration response sequence under preset working conditions is matched with the features of the hierarchical reference samples, the model can directly and quickly identify the fault type and degree based on the trained hierarchical features, avoiding blind feature comparison without a benchmark, significantly shortening the identification cycle, and thus improving the efficiency of clutch friction element deformation fault identification.

[0060] For example, three reference friction elements are pre-selected. The first reference friction element is chosen, and eight measuring points are evenly selected along its inner edge circumferentially. A dial indicator is used to measure the axial warpage height of each measuring point relative to the outer edge reference surface. The measured data are 0.01mm, 0.012mm, 0.009mm, 0.011mm, 0.013mm, 0.008mm, 0.014mm, and 0.01mm, respectively. The average value is calculated as (0.01+0.012+0.009+0.011+0.013+0.008+0.01). (4 + 0.01) ÷ 8 = 0.011 mm. The first preset range is set to 0~0.03 mm. This deformation quantification index of 0.011 mm falls within the first preset range and is judged as a healthy benchmark sample. A second reference friction element is selected, and eight measurement points are evenly selected along its inner edge circumferentially. A dial indicator is used to measure the axial warping height of each measurement point relative to the outer edge reference surface. The measured data are 0.035 mm, 0.04 mm, 0.032 mm, 0.038 mm, 0.042 mm, 0.036 mm, and 0.039 mm, respectively. The deformation is 0.041 mm. The calculated average value is (0.035+0.04+0.032+0.038+0.042+0.036+0.039+0.041)÷8=0.038 mm. The second preset range is set to 0.03~0.1 mm. The deformation quantification index of 0.038 mm is within the second preset range and is judged as a micro-deformation reference sample. A third reference friction element is selected, and eight measurement points are evenly selected along the circumferential direction on its inner edge. The axial warpage height of each measurement point relative to the outer edge reference surface is measured using a dial indicator. The measured data were 0.12mm, 0.11mm, 0.13mm, 0.125mm, 0.115mm, 0.135mm, 0.122mm, and 0.118mm, respectively. The average value was calculated as (0.12+0.11+0.13+0.125+0.115+0.135+0.122+0.118)÷8=0.122mm. The third preset interval was set to be greater than or equal to 0.1mm. The deformation quantification index of 0.122mm was within the third preset interval and was judged as a severely deformed benchmark sample.

[0061] In some embodiments, acquiring the operating vibration response sequence of the target friction element under preset operating conditions includes: constructing a test bench for clutch disengagement conditions and installing the target friction element inside the clutch housing; wherein, when installing the target friction element inside the clutch housing, considering the differences in constraint conditions at different positions inside the clutch, the position near the snap ring inside the clutch housing is determined to be a deformation-sensitive position; the target friction element is assembled at the deformation-sensitive position to ensure that the maximum amplitude of abnormal hydrodynamic vibration caused by warping deformation can be captured under disengagement conditions; the brake is controlled to lock the steel plate, and the input shaft is driven to rotate the friction plate, so that a relative speed difference is generated between the friction plate and the steel plate; when the input shaft speed stabilizes at the speed value corresponding to the preset operating condition, the vibration acquisition system is triggered; the vibration signal of the clutch housing is synchronously acquired using orthogonally arranged sensors, and it is determined whether the acquisition time has reached the preset sample window length. If it has, the acquisition is stopped and the operating vibration response sequence is output. According to embodiments of the present invention, by selecting a deformation-sensitive location to assemble the target friction element, the maximum amplitude hydrodynamic abnormal vibration caused by warping deformation can be captured by utilizing the difference in constraint conditions at that location, thereby enhancing the intensity of fault characteristic signals. A separate working condition test bench is used, and the brake is controlled to lock the steel plate, driving the input shaft to form a relative speed difference, simulating the actual fault-induced working condition to ensure effective manifestation of fault characteristics. Data acquisition is triggered when the input shaft speed stabilizes at a preset value, avoiding interference from unstable working conditions and ensuring that the acquired signals are valid data under the target working condition. Vibration signals are synchronously acquired using orthogonally arranged sensors, enabling multi-dimensional capture of fault characteristics and reducing signal omissions. Vibration response sequences are acquired and output according to a preset sample window length, ensuring data integrity and consistency, facilitating subsequent rapid analysis and identification, and thus improving the efficiency of clutch friction element deformation fault identification.

[0062] S102, the running vibration response sequence is converted into a frequency domain signal to obtain the original response spectrum;

[0063] S103, Based on the pre-constructed frequency band discrimination evaluation model, determine the adaptive filtering weights of each frequency band in the original response spectrum; wherein, the construction logic of the frequency band discrimination evaluation model is as follows: analyze the spectral characteristics of each reference sample under different deformation states; if a frequency band exhibits high homogeneity characteristics among different deformation states, then the frequency band is determined to be a noise-dominant frequency band, and its corresponding adaptive filtering weight is determined to be a suppression coefficient; if a frequency band exhibits low homogeneity characteristics among different deformation states, then the frequency band is determined to be a feature-sensitive frequency band, and its corresponding adaptive filtering weight is determined to be a retention coefficient.

[0064] In some embodiments, analyzing the spectral characteristics of each reference sample under different deformation states includes: calculating the spectral vector distance and spectral vector deflection angle of any two sets of reference samples under different deformation states in the same frequency band; synthesizing the inter-state difference score of the frequency band based on the spectral vector distance and the spectral vector deflection angle; if the inter-state difference score is less than a preset homogenization judgment threshold, it indicates that the signal characteristics in the frequency band tend to be consistent between the two sets of states, and the frequency band is judged to exhibit the high homogenization characteristic; if the inter-state difference score is greater than or equal to the homogenization judgment threshold, it indicates that the signal characteristics in the frequency band have significant distinguishability, and the frequency band is judged to exhibit the low homogenization characteristic. According to embodiments of the present invention, the feature differences of reference samples in different deformation states in the same frequency band are quantified by two dimensions: probability density divergence value and spectral vector deflection angle value, so as to achieve accurate quantification of differences and avoid the limitations of a single indicator. Based on the difference score between the synthesized states of the two types of quantified values, a standardized difference judgment criterion is established to improve the objectivity of feature differentiation. A homogenization judgment threshold is introduced to screen out frequency bands with low homogenity and high distinguishability, and eliminate frequency bands with high homogenity and no distinguishing value, thereby reducing the interference of invalid feature data, focusing on core effective features, and thus improving the efficiency of clutch friction element deformation fault identification.

[0065] In some embodiments, state difference scoring The calculation expression is:

[0066]

[0067] in, The distance between the two spectral vectors; The spectral vector deflection angle is used to characterize the frequency band. The two spectral vectors within and The angle between them.

[0068] In some embodiments, The calculation expression is:

[0069]

[0070] Where M is the mean of the selected spectrum and V is the variance of the corresponding spectrum.

[0071] In some embodiments, The calculation expression is:

[0072] .

[0073] In some embodiments, the frequency band discrimination evaluation model further includes global difference weighting logic: for cases with K different deformation states, calculate the state difference score of the m-th frequency band under all possible state combinations; sum all state difference scores with weights to obtain the global discrimination total score of the m-th frequency band; if the global discrimination total score is in the low quantile range of all frequency band score sequences, it is confirmed that the frequency band cannot effectively represent deformation differences, and its corresponding adaptive filtering weight is determined to be zero or close to zero, so as to remove the influence of the frequency band in subsequent steps. According to the embodiments of the present invention, by eliminating invalid frequency bands that cannot effectively represent deformation differences, the interference of invalid data on the identification process is reduced, and the amount of data processing is reduced; the discrimination of each frequency band is quantified based on global difference weighting logic, and effective frequency bands sensitive to different deformation states are accurately selected, improving the targeting of feature dimensions; the noise interference introduced by invalid frequency bands is avoided, the signal-to-noise ratio of effective features is improved, and the fault identification model is made more focused on key difference information, thereby improving the efficiency of clutch friction element deformation fault identification.

[0074] For example, let K = 3 be the number of deformation state types (corresponding to healthy baseline samples, slightly deformed baseline samples, and severely deformed baseline samples), and select a specific frequency band as the m-th frequency band (e.g., the 500Hz frequency band). There are 3 possible combinations of the 3 deformation states: Combination 1: Healthy baseline sample vs. slightly deformed baseline sample; Combination 2: Healthy baseline sample vs. severely deformed baseline sample; Combination 3: Slightly deformed baseline sample vs. severely deformed baseline sample. For Combination 1, the spectral vector distance JMDistance1 = 0.15, and the spectral vector deflection angle γ1 = 30°, so the state difference score JMScore1 = 0.15 × tan30° = 0.087. For Combination 2, the spectral vector distance JMDistance2 = 0.42, and the spectral vector deflection angle γ2 = 60°, so JMScore2 = 0.42 × tan60° = 0.727. For Combination 3, the spectral vector distance JMDistance3 = 0.31, and the spectral vector difference score JMScore1 = 0.15 × tan30° = 0.087. Given a deflection angle γ3 = 45°, JMScore3 = 0.31 × tan45° = 0.31; assuming all three combinations have a weight of 1 (equal weighting), the total global discrimination score for the m-th frequency band is 0.087 + 0.727 + 0.31 = 1.124; assuming the total global discrimination score sequence for all frequency bands (100 bands in total) is [0.23, 0.35, ..., 1.124, ..., 3.87], sorted in ascending order... Then, the low quantile interval is set as the top 20% (i.e., score ≤ 0.56). Since the global discrimination score of the m-th frequency band is 1.124 > 0.56, it is not in the low quantile interval, so the frequency band is retained and the adaptive filtering weight is set to 1 (retention coefficient). If another frequency band (the n-th frequency band) has a global discrimination score of 0.41, which is in the low quantile interval of the top 20%, then its adaptive filtering weight is determined to be 0.001 (approaching zero), and the influence of this frequency band is eliminated in subsequent steps.

[0075] S104, The original response spectrum is weighted and corrected using the adaptive filtering weights to obtain the feature-enhanced spectrum;

[0076] In some embodiments, the weighted correction of the original response spectrum using the adaptive filtering weights includes: constructing a frequency domain mask vector that corresponds one-to-one with the frequency points of the original response spectrum, wherein the element values ​​in the frequency domain mask vector are the adaptive filtering weights; performing point-by-point multiplication of the original response spectrum with the frequency domain mask vector; if the adaptive filtering weight corresponding to a certain frequency point is the suppression coefficient, then the amplitude of that frequency point is attenuated, thereby suppressing background noise interference at that frequency point in the generated feature-enhanced spectrum; if the adaptive filtering weight corresponding to a certain frequency point is the retention coefficient, then the amplitude of that frequency point is retained or amplified, thereby highlighting the vibration characteristics related to deformation faults. According to embodiments of the present invention, by constructing frequency domain mask vectors that correspond one-to-one with the frequency points of the original response spectrum, precise weighted correction of the target frequency points is achieved through point-by-point multiplication, avoiding misprocessing of irrelevant frequency points; a suppression coefficient is applied to the frequency points corresponding to background noise to directly attenuate their amplitude, effectively reducing the interference of background noise on fault characteristics; a retention coefficient is applied to the frequency points related to deformation faults to maintain or amplify their amplitude, thereby achieving targeted enhancement of fault vibration characteristics and improving the efficiency of clutch friction element deformation fault identification.

[0077] For example, the original response spectrum contains five frequency points: 100Hz, 300Hz, 500Hz, 700Hz, and 900Hz, with amplitudes of 0.8V, 1.2V, 0.9V, 1.5V, and 0.7V respectively. Therefore, the original response spectrum can be represented as [0.8, 1.2, 0.9, 1.5, 0.7]. A frequency domain mask vector is constructed, corresponding one-to-one with each frequency point of the original response spectrum. The suppression coefficient is set to 0.2, and the retention coefficient to 1.5. The adaptive filtering weights for each frequency point are determined according to the frequency band discrimination evaluation model, resulting in a frequency domain mask vector of [0.2, 1.5, 1.5, 0.2, 1.5]. The original response spectrum and the frequency domain mask vector are then compared. The values ​​are multiplied point by point, specifically: 100Hz frequency point: 0.8V×0.2=0.16V (amplitude attenuation, suppressing background noise), 300Hz frequency point: 1.2V×1.5=1.8V (amplitude amplification, highlighting fault characteristics), 500Hz frequency point: 0.9V×1.5=1.35V (amplitude amplification, highlighting fault characteristics), 700Hz frequency point: 1.5V×0.2=0.3V (amplitude attenuation, suppressing background noise), 900Hz frequency point: 0.7V×1.5=1.05V (amplitude amplification, highlighting fault characteristics); the final feature enhancement spectrum is [0.16,1.8,1.35,0.3,1.05].

[0078] S105, Calculate the spectral energy disorder index based on the energy distribution probability of the enhanced spectrum.

[0079] In some embodiments, calculating the spectral energy disorder index includes: normalizing the enhanced spectrum to obtain the relative probability proportion of each frequency point amplitude in the total energy; calculating the complexity measure of the enhanced spectrum using an information entropy algorithm based on the relative probability proportion; if the calculated complexity measure value is greater than a preset measure threshold, the energy distribution state of the enhanced spectrum is determined to be chaotic, and the higher the corresponding spectral energy disorder index, the more significant the nonlinear influence of the deformation degree of the friction element on the system dynamic characteristics. According to embodiments of the present invention, by normalizing the enhanced spectrum to convert the amplitude of each frequency point into a relative probability proportion under the total energy, the interference of the absolute magnitude of the amplitude is eliminated, focusing on the energy distribution law and improving the effectiveness of the feature; the introduction of the information entropy algorithm to quantify the complexity measure of the spectrum establishes a direct correlation between the degree of energy distribution chaos and the spectral energy disorder index, realizing a quantitative characterization of the nonlinear influence caused by the deformation of the friction element; by directly mapping the level of the spectral energy disorder index to the degree of influence of the friction element deformation on the system dynamic characteristics, complex multi-dimensional parameter fitting is not required, simplifying the steps of fault correlation analysis, thereby improving the efficiency of clutch friction element deformation fault identification.

[0080] For example, the feature-enhanced spectrum contains 5 frequency points, with amplitudes of 0.16V at 100Hz, 1.8V at 300Hz, 1.35V at 500Hz, 0.3V at 700Hz, and 1.05V at 900Hz; the total energy of the feature-enhanced spectrum is 0.16 + 1.8 + 1.35 + 0.3 + 1.05 = 4.66V; after normalization, the relative probability of the 100Hz frequency point is approximately 0.16 ÷ 4.66 ≈ 0.0343; the relative probability of the 300Hz frequency point is approximately 1.8 ÷ 4.66 ≈ 0.3863; the relative probability of the 500Hz frequency point is approximately 1.35 ÷ 4.66 ≈ 0.2897; the relative probability of the 700Hz frequency point is approximately 0.3 ÷ 4.66 ≈ 0.0644; and the relative probability of the 900Hz frequency point is approximately 1.05 ÷ 4.66 ≈ 0.2253. The information entropy algorithm is used. The computational complexity measures are: 0.0343×ln0.0343≈-0.1159, 0.3863×ln0.3863≈-0.3666, 0.2897×ln0.2897≈-0.3604, 0.0644×ln0.0644≈-0.1758, 0.2253×ln0.2253≈0.2253×(-1.489)≈-0.3355; summing and taking the negative gives -[(-0.1159)+(-0.3666)+(-0.3604)+(-0.1758)+(-0.3355)] = -(-1.3542) = 1.3542; The preset metric threshold is set to 1.0. Since the calculated complexity metric 1.3542 > 1.0, the energy distribution state of the feature enhancement spectrum is determined to be chaotic.

[0081] S106, input the spectral energy disorder index into the fault identification classifier to determine the deformation fault category of the target friction element.

[0082] In some embodiments, the step of inputting the spectral energy disorder index into a fault identification classifier to determine the deformation fault category of the target friction element includes: inputting the spectral energy disorder index of a reference sample with known deformation fault categories as a training set into the classification model; optimizing the decision boundary of the classification model using the training set so that it can distinguish entropy value intervals corresponding to different degrees of deformation; mapping the spectral energy disorder index of the target friction element into the decision boundary; determining the region to which the spectral energy disorder index falls; if it falls into a first region, outputting the identification result of the health status; if it falls into a second region, outputting the deformation fault and its corresponding level label. According to embodiments of the present invention, by training a classifier based on the spectral energy disorder index of known fault category samples, the model learns the correspondence between fault categories and feature parameters, establishing an accurate fault discrimination benchmark; the decision boundary of the classification model is optimized, clarifying the entropy value range corresponding to different degrees of deformation, realizing quantitative differentiation of fault categories, and avoiding fuzzy judgment; the spectral energy disorder index of the target component is directly mapped to the optimized decision boundary, and the result is directly output through region assignment, omitting complex feature analysis and manual judgment steps, thus improving the efficiency of the identification process; integrated judgment of health status, fault status, and fault level is realized, eliminating the need for step-by-step detection, shortening the overall identification cycle, and thereby improving the efficiency of clutch friction element deformation fault identification.

[0083] For example, select 50 reference samples from each of three known deformation categories: 50 healthy baseline samples, 50 slightly deformed baseline samples, and 50 severely deformed baseline samples. Calculate the spectral energy disorder index for each group of samples to obtain the training set data. The spectral energy disorder index ranges from 0.3 to 0.6 for healthy samples, from 0.8 to 1.2 for slightly deformed samples, and from 1.5 to 2.0 for severely deformed samples. Input the above training set into a support vector machine classification model and optimize to obtain the decision boundary. The first region is where the spectral energy disorder index is ≤0.7, and the second region includes two... The system is divided into several sub-regions, where the second sub-region 1 has a spectral energy disorder index of 0.7 < 1.3 and the second sub-region 2 has a spectral energy disorder index > 1.3. The spectral energy disorder index of the target friction element is calculated to be 0.52, and this index is mapped to the decision boundary. If 0.52 falls into the first region, the health status identification result is output. If the spectral energy disorder index of the target friction element is 1.05, it is determined to fall into the second sub-region 1, and the micro-deformation fault level label is output. If the index is 1.83, it is determined to fall into the second sub-region 2, and the severe deformation fault level label is output.

[0084] In some embodiments, the training set uses 5-fold cross-validation to optimize model parameters and avoid overfitting; the decision boundary is optimized using a grid search algorithm with a search range of spectral disorder index of 0-2.5 and a boundary error of ≤0.01; the division of the first and second regions is based on the entropy distribution of the training set samples, and the K-means clustering algorithm is used to determine the interval boundaries to ensure the rationality and stability of the interval division.

[0085] In some embodiments, the fault identification classifier adopts a support vector machine (SVM) classification model, the kernel function is selected as the radial basis function (RBF), the penalty coefficient C=10, and the gamma parameter=0.1.

[0086] According to embodiments of the present invention, the present invention also provides an electronic device and a readable storage medium.

[0087] Figure 2 A schematic block diagram of an electronic device that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0088] The electronic device includes a computing unit 201, which can perform various appropriate actions and processes based on a computer program stored in ROM 202 or a computer program loaded into RAM 203 from storage unit 208. RAM 203 can also store various programs and data required for the operation of the electronic device. The computing unit 201, ROM 202, and RAM 203 are interconnected via bus 204. I / O interface 205 is also connected to bus 204.

[0089] Multiple components in the electronic device are connected to the I / O interface 205, including: an input unit 206, such as a keyboard, mouse, etc.; an output unit 207, such as various types of displays, speakers, etc.; a storage unit 208, such as a disk, optical disk, etc.; and a communication unit 209, such as a network card, modem, wireless transceiver, etc. The communication unit 209 allows the electronic device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0090] The computing unit 201 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 201 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 201 performs the various methods and processes described above, such as the clutch friction element deformation fault identification method. For example, in some embodiments, the clutch friction element deformation fault identification method can be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 208. In some embodiments, part or all of the computer program can be loaded and / or installed on an electronic device via ROM 202 and / or communication unit 209. When the computer program is loaded into RAM 203 and executed by the computing unit 201, one or more steps of the clutch friction element deformation fault identification method described above can be performed. Alternatively, in other embodiments, the computing unit 201 can be configured to perform the clutch friction element deformation fault identification method by any other suitable means (e.g., by means of firmware).

[0091] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0092] The program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This program code can be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0093] In the context of this invention, a readable storage medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A readable storage medium can be a machine-readable signal medium or a machine-readable storage medium. A readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0094] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including voice input, speech input, or tactile input).

[0095] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0096] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0097] It should be understood that the various processes described above can be used to rearrange, add, or delete steps. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0098] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for identifying deformation faults in clutch friction components, characterized in that, include: Obtain the vibration response sequence of the target friction element under preset operating conditions; The vibration response sequence is converted into a frequency domain signal to obtain the original response spectrum; Based on a pre-constructed frequency band discrimination evaluation model, the adaptive filtering weights of each frequency band in the original response spectrum are determined. The construction logic of the frequency band discrimination evaluation model is as follows: Analyze the spectral characteristics of each reference sample under different deformation states; if a frequency band exhibits high homogeneity across different deformation states, it is determined to be a noise-dominant frequency band, and its corresponding adaptive filtering weight is determined as a suppression coefficient; if a frequency band exhibits low homogeneity across different deformation states, it is determined to be a feature-sensitive frequency band, and its corresponding adaptive filtering weight is determined as a retention coefficient. The original response spectrum is weighted and corrected using the adaptive filtering weights to obtain the feature-enhanced spectrum; Based on the energy distribution probability of the enhanced spectrum, the spectral energy disorder index is calculated. The spectral energy disorder index is input into the fault identification classifier to determine the deformation fault category of the target friction element.

2. The method for identifying deformation faults in clutch friction elements according to claim 1, characterized in that, The analysis of the spectral characteristics of each reference sample under different deformation states includes: Calculate the spectral vector distance and spectral vector deflection angle between any two sets of reference samples in different deformation states in the same frequency band; Based on the spectral vector distance and the spectral vector deflection angle, a state difference score for this frequency band is synthesized. If the difference score between states is less than the preset homogeneity judgment threshold, it indicates that the signal characteristics of the frequency band tend to be consistent between the two sets of states, and the frequency band is judged to exhibit the high homogeneity characteristics. If the difference score between states is greater than or equal to the homogenization determination threshold, it indicates that the signal features in this frequency band have significant distinguishability, and the frequency band is determined to exhibit the low homogenization feature.

3. The method for identifying clutch friction element deformation faults according to claim 2, characterized in that, The frequency band discrimination evaluation model also includes global difference weighting logic: For cases with K different deformation states, calculate the state difference score for the m-th frequency band under all possible state combinations; The total global discrimination score for the m-th frequency band is obtained by weighted summation of all state difference scores. If the total global discrimination score is located in the low quantile range of all frequency band score sequences, it is confirmed that the frequency band cannot effectively represent the deformation difference, and its corresponding adaptive filtering weight is determined to be zero or close to zero, so as to remove the influence of the frequency band in subsequent steps.

4. The method for identifying clutch friction element deformation faults according to claim 3, characterized in that, The step of weighting and correcting the original response spectrum using the adaptive filtering weights includes: Construct a frequency domain mask vector that corresponds one-to-one with the frequency points of the original response spectrum, and the element values ​​in the frequency domain mask vector are the adaptive filtering weights; Perform point-by-point multiplication between 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, then the amplitude of that frequency point is attenuated, thereby suppressing the background noise interference of that frequency point in the generated feature enhancement spectrum; If the adaptive filtering weight corresponding to a certain frequency point is the retention coefficient, then the amplitude of that frequency point is maintained or amplified, thereby highlighting the vibration characteristics related to deformation faults.

5. The method for identifying deformation faults in clutch friction elements according to claim 1, characterized in that, The calculation of the spectral energy disorder index includes: The enhanced spectrum of the feature is normalized to obtain the relative probability of each frequency point amplitude in the total energy; Based on the relative probability proportions, the complexity measure of the feature enhancement spectrum is calculated using the information entropy algorithm; If the calculated complexity metric value is greater than the preset metric threshold, the energy distribution state of the feature-enhanced spectrum is determined to be chaotic. The higher the corresponding spectral disorder index, the more significant the nonlinear influence of the deformation degree of the friction element on the system dynamic characteristics.

6. The method for identifying deformation faults in clutch friction elements according to claim 1, characterized in that, Before obtaining the vibration response sequence of the target friction element under preset operating conditions, the process also includes establishing a reference sample deformation level, specifically: Several measurement points are uniformly selected along the circumferential direction on the inner edge of the pre-selected reference friction element, and the axial warping height of each measurement point relative to the outer edge reference surface is measured. Calculate the average axial warping height of all measurement points as a deformation quantification index; If the deformation quantification index is within the first preset range, the friction element is determined to be a healthy benchmark sample. If the deformation quantification index is within the second preset range, then the friction element is determined to be a micro-deformation benchmark sample; If the deformation quantification index is within the third preset range, the friction element is determined to be a severely deformed benchmark sample. The healthy baseline samples, the slightly deformed baseline samples, and the severely deformed baseline samples together constitute the training dataset used to construct the frequency band discrimination evaluation model.

7. The method for identifying deformation faults in clutch friction elements according to claim 1, characterized in that, The acquisition of the operating vibration response sequence of the target friction element under preset working conditions includes: A test bench for clutch disengagement conditions was constructed, and the target friction element was installed inside the clutch housing. When installing the target friction element inside the clutch housing, considering the differences in constraint conditions at different locations inside the clutch, the position near the snap ring inside the clutch housing was determined to be a deformation-sensitive position. The target friction element was assembled at the deformation-sensitive position to ensure that the maximum amplitude of abnormal hydrodynamic vibration caused by warping deformation could be captured under disengagement conditions. The control brake locks the steel plate and drives the input shaft to rotate the friction plate, creating a relative speed difference between the friction plate and the steel plate; When the input shaft speed stabilizes at the speed value corresponding to the preset working condition, the vibration acquisition system is triggered; Vibration signals of the clutch housing are synchronously acquired using orthogonally arranged sensors. The acquisition time is then determined to be within the preset sample window length. If the acquisition time is reached, the acquisition is stopped and the operating vibration response sequence is output.

8. The method for identifying deformation faults in clutch friction elements according to claim 1, characterized in that, The step of inputting the spectral energy disorder index into the fault identification classifier to determine the deformation fault category of the target friction element includes: The spectral energy disorder index of reference samples with known deformation fault categories is used as the training set input to the classification model. The decision boundary of the classification model is optimized using the training set so that it can distinguish the entropy value range corresponding to different degrees of deformation. The spectral energy disorder index of the target friction element is mapped to the decision boundary; Determine the region to which the spectral disorder index falls. If it falls into the first region, output the health status identification result; if it falls into the second region, output the deformation fault and its corresponding level label.

9. An electronic device, characterized in that, include: At least one processor; A memory that is communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, enables the at least one processor to perform the method described in any one of claims 1-8.

10. A non-transitory computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-8.

Images