Operation fault monitoring method and system applied to automobile transmission system

Abstract

The invention relates to the technical field of transmission system fault diagnosis, in particular to an operation fault monitoring method and system applied to an automobile transmission system. Determining the vibration characteristic coefficient of each period according to the amplitude difference of the vibration sequence in different frequency regions on the frequency domain; and a fault trend coefficient is constructed based on the data fluctuation frequency of the vibration signals of all periods and the change trend of the vibration characteristic coefficients of all periods so as to adjust the abnormal proportion, and based on the adjusted abnormal proportion, the operation fault monitoring of the automobile transmission system is carried out in combination with an isolated forest algorithm. The method improves the sensitivity of abnormal detection at the initial stage of a fault or during a slight fault, guarantees the dynamic balance of the false alarm rate and the missing report rate of the fault detection of the transmission system, and improves the quality of the fault detection of the automobile transmission system.

Description

Technical Field

[0001] The present application relates to the technical field of transmission system fault diagnosis, and particularly to an operation fault monitoring method and system for an automotive transmission system. Background Art

[0002] With the rapid development of the modern automotive industry, the complexity and automation level of vehicles are continuously increasing, and it has become crucial to ensure the stability and reliability of the automotive transmission system. As the core component connecting the engine and the wheels, the transmission system not only determines the power transmission efficiency of the vehicle but also directly affects the driving experience and safety. Facing the growing safety requirements and higher expectations for vehicle performance, detecting and predicting potential faults in the transmission system in a timely and accurate manner has become an important task for manufacturers and service providers. With the development of intelligent monitoring technology, real-time online monitoring technology has promoted the development of fault monitoring towards intelligence and adaptability.

[0003] However, the existing fault monitoring solutions for automotive transmission systems mainly focus on the instantaneous feature analysis of vibration signals. After a fault occurs, the frequency-domain features of the instantaneous vibration signals are analyzed. However, it is often after obvious faults have occurred that the diagnosis of the faults is realized through the frequency-domain features of the vibration signals. At this time, certain wear has already occurred to other components in the automotive transmission system, reducing the service life of the transmission system. In the case of early or minor faults, it is difficult to identify the faults only through this diagnostic method that focuses on instantaneous features. Summary of the Invention

[0004] In order to solve the above technical problems, the purpose of the present application is to provide an operation fault monitoring method and system for an automotive transmission system, and the specific technical solutions adopted are as follows:

[0005] In the first aspect, an embodiment of the present application provides an operation fault monitoring method for an automotive transmission system, and the method includes the following steps:

[0006] Collect all vibration signals within each cycle during the operation of the automotive transmission system, and construct a vibration sequence for each cycle;

[0007] Based on the periodic change characteristics of the elements in the vibration sequence of each cycle, the fluctuation condition of the vibration sequence, and the amplitude difference between different frequency regions of the vibration sequence in the frequency domain, determine the vibration characteristic coefficient of each cycle;

[0008] Based on the vibration sequences of all cycles, construct a total vibration sequence; based on the vibration characteristic coefficients of all cycles, construct a vibration characteristic sequence; based on the frequency of data fluctuations in the total vibration sequence and the data change trend in the vibration characteristic sequence, construct a fault trend coefficient for the automotive transmission system;

[0009] Determine the abnormal proportion adjustment value of the automotive drive system based on the fault trend coefficient;

[0010] Based on the abnormal proportion adjustment value, combine the isolation forest algorithm to monitor the operation faults of the automotive drive system.

[0011] In one embodiment, the acquisition process of the vibration characteristic coefficients of each period is as follows:

[0012] Obtain the autocorrelation coefficients of the vibration sequences of each period through the autocorrelation function, and construct the autocorrelation function graph of the vibration sequences of each period; based on the peak value distribution characteristics in the autocorrelation function graph, construct the period characteristic values of the autocorrelation function graph of each period;

[0013] Calculate the variance of all elements in the vibration sequence of each period, denoted as the first variance;

[0014] Obtain the frequency-domain waveform diagram of the vibration sequence of each period through the time-frequency conversion algorithm. In each frequency-domain waveform diagram, divide all frequencies to obtain the low-frequency region, the middle-frequency region, and the high-frequency region; calculate the average value of the variances of the amplitudes of all frequencies in the middle-frequency region and the high-frequency region, denoted as the first average value; record the frequencies with amplitudes greater than the preset frequency significance threshold in the frequency-domain waveform diagram as significant frequencies, and record the sequence composed of the amplitudes of all significant frequencies in the low-frequency region as the low-frequency amplitude sequence; obtain the low-frequency distribution characteristic measure of the low-frequency region based on the element changes in the first-order difference sequence of the low-frequency amplitude sequence;

[0015] Based on the period characteristic value and the first variance, combine the difference between the first average value and the low-frequency distribution characteristic measure to construct the vibration characteristic coefficient of each period.

[0016] In one embodiment, the acquisition process of the period characteristic value is as follows:

[0017] Record the peaks with peak values greater than the preset correlation threshold in the autocorrelation function graph of each period as significant peaks;

[0018] Take the lag amount corresponding to the peak value of each significant peak as the ordinate and the order of appearance of each significant peak as the abscissa to obtain the coordinate points corresponding to each significant peak; fit the coordinate points corresponding to all significant peaks in the autocorrelation function graph through the linear fitting algorithm, and record the obtained fitting line as the lag amount fitting line; calculate the determination coefficient of the lag amount fitting line;

[0019] Take the product of the mean value of the peak values of all significant peaks in the autocorrelation function graph of each period and the determination coefficient as the period characteristic value of the autocorrelation function graph of each period.

[0020] In one embodiment, the expression of the low-frequency distribution characteristic measure is:

[0021] Wherein, lf i is the low-frequency distribution characteristic measure in the low-frequency region of the frequency-domain waveform diagram in the i-th cycle, μ(L i ), σ(L i ) are respectively the mean and variance of all elements in the first-order difference sequence of the low-frequency amplitude sequence in the i-th cycle, and norm() is a normalization function.

[0022] In one embodiment, the expression of the vibration characteristic coefficient is:

[0023] Wherein, A i is the vibration characteristic coefficient in the i-th cycle, R i is the cycle eigenvalue in the i-th cycle, σ(V i ) is the first variance in the i-th cycle, t i is the lag amount corresponding to the peak value of the first significant peak in the autocorrelation function diagram in the i-th cycle, mf i is the first average value in the i-th cycle, lf i is the low-frequency distribution characteristic measure in the i-th cycle, and exp() is an exponential function with the natural constant e as the base.

[0024] In one embodiment, the process of obtaining the fault trend coefficient of the automotive transmission system is as follows:

[0025] Take the total vibration sequence as the input of the polynomial fitting algorithm, and the output is the fitting curve of the total vibration sequence;

[0026] Use the time series decomposition algorithm to obtain the trend sequence of the vibration characteristic sequence;

[0027] Construct the fault trend coefficient of the automotive transmission system based on the change of the fitting curve and the fitting straight line of the trend sequence.

[0028] In one embodiment, the expression of the fault trend coefficient is:

[0029] B = lg(X) × exp(-K), where B is the fault trend coefficient of the automotive transmission system, X is the number of data points with a derivative of 0 on the fitting curve, K is the slope of the fitting straight line of the trend sequence, lg() is a logarithmic function with 10 as the true number, and exp() is an exponential function with the natural constant e as the base.

[0030] In one embodiment, the expression of the abnormal ratio adjustment value of the automotive transmission system is:

[0031] Wherein, W is the abnormal ratio adjustment value of the vehicle transmission system, B is the fault trend coefficient of the vehicle transmission system, and arctan() is the arctangent function.

[0032] In one embodiment, based on the abnormal ratio adjustment value, the vehicle transmission system operation fault monitoring is carried out in combination with the Isolation Forest algorithm, specifically as follows:

[0033] Taking the vibration sequences of all cycles as the input of the Isolation Forest algorithm, taking the abnormal ratio adjustment value as the abnormal ratio in the Isolation Forest algorithm, and the output is the abnormal score of each cycle vibration sequence; obtaining the segmentation threshold of the abnormal scores of the vibration sequences of all cycles;

[0034] Taking the cycles with the abnormal scores of the vibration sequences greater than the segmentation threshold as the abnormal cycles; obtaining the sequence composed of all moments corresponding to all the abnormal cycles and recording it as the abnormal sequence, calculating the Hurst exponent of the first-order difference sequence of the abnormal sequence, if the Hurst exponent is greater than the preset fault threshold, then there is an operation fault in the current vehicle transmission system; otherwise, the current vehicle transmission system is operating normally.

[0035] In a second aspect, the embodiments of the present application also provide an operation fault monitoring system applied to a vehicle transmission system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the steps of the method described in any one of the above are implemented.

[0036] The embodiments of the present application at least have the following beneficial effects:

[0037] According to the time-frequency domain characteristics of the instantaneous vibration data of each cycle during the operation of the vehicle transmission system, the present application calculates the vibration characteristic coefficient, which helps to improve the accuracy of identifying the difference between normal and abnormal states; then, in combination with the time-series change trend of the vibration data and the vibration characteristic coefficient, the fault trend coefficient is calculated, which helps to improve the accuracy and sensitivity of subsequent fault diagnosis; finally, the abnormal ratio is adaptively adjusted according to the fault trend coefficient, balancing the accuracy and sensitivity of the Isolation Forest anomaly detection. In this way, by combining the time-frequency domain characteristics and the time-series change trend of the vibration data, the sensitivity of anomaly detection in the initial stage of the fault or in the case of minor faults is improved, and it can respond flexibly according to the actual vibration situation, improving the adaptability and robustness of anomaly detection, ensuring the dynamic balance of the false alarm rate and the missed alarm rate of the transmission system fault detection, improving the quality of the vehicle transmission system fault detection, thereby realizing the high-precision and high-sensitivity operation fault monitoring of the vehicle transmission system, ensuring the safety of the vehicle transmission system and extending the service life to a certain extent. Description of the Drawings

[0038] To more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0039] Figure 1 The flowchart of the steps of a method for monitoring the operating faults of an automotive transmission system provided by an embodiment of the present application;

[0040] Figure 2 The schematic diagram of the frequency-domain waveform diagram under normal conditions;

[0041] Figure 3 The schematic diagram of the frequency-domain waveform diagram under minor faults;

[0042] Figure 4 The schematic diagram of the frequency-domain waveform diagram under severe faults;

[0043] Figure 5 The schematic diagram of the autocorrelation function diagram. Detailed implementation manners

[0044] In order to further elaborate on the technical means and effects adopted by the present application to achieve the intended invention purpose, the following, in combination with the drawings and preferred embodiments, details the specific implementation manners, structures, features, and effects of the method and system for monitoring the operating faults of an automotive transmission system proposed according to the present application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs.

[0046] The following specifically describes the specific solutions of the method and system for monitoring the operating faults of an automotive transmission system provided by the present application with reference to the drawings.

[0047] Please refer to Figure 1 , which shows the flowchart of the steps of a method for monitoring the operating faults of an automotive transmission system provided by an embodiment of the present application. The method includes the following steps:

[0048] Step S1, collect all vibration signals in each cycle during the operation of the automotive transmission system, and construct the vibration sequence of each cycle.

[0049] In this embodiment, a vibration sensor is installed on the gearbox housing of the automotive transmission system to collect the vibration signals during the operation of the transmission system. Specifically:

[0050] When collecting vibration signals, the signal acquisition frequency is set to 5200 Hz, the vibration signal acquisition duration for each time is set to 1 second, that is, the acquisition period is 1 second, and the sequence composed of all the vibration signals collected within the acquisition period is used as the vibration sequence for each acquisition period.

[0051] It should be noted that for the setting of the acquisition frequency and acquisition duration when collecting vibration signals each time, this application only provides one setting method. Implementers can set the acquisition frequency and acquisition duration for each acquisition according to the actual situation, and this application does not make specific restrictions.

[0052] Taking the vibration sequences of each period as the input of the wavelet denoising algorithm, the collected vibration signals are denoised to obtain the denoised vibration sequences. Among them, the wavelet denoising algorithm is a well-known technology, and the specific process will not be elaborated here. It should be noted that for the denoising of vibration signals, this application only provides one denoising method. There are many existing denoising methods, and implementers can also use other denoising algorithms to denoise the vibration signals, and this application does not make specific restrictions.

[0053] Step S2: Based on the periodic change characteristics of the elements in the vibration sequences of each period, the fluctuation conditions of the vibration sequences, and the amplitude differences between different frequency regions of the vibration sequences in the frequency domain, determine the vibration characteristic coefficients of each period.

[0054] In the operation fault detection of the automotive transmission system, the existing detection method is to analyze the time-frequency domain of the collected data by observing the changes in vibration data during operation to determine whether there is an operation fault. However, this method focuses on the instantaneous changes of vibration data, which means that only when a relatively significant fault occurs will the time-frequency domain characteristics of the vibration data change significantly, so that it can be determined whether there is a fault. However, in the initial stage of the fault or when the fault is relatively minor, the time-frequency domain change of the vibration data is relatively slight, making it difficult to diagnose based only on the instantaneous change situation. By the time it can be diagnosed, it means that a relatively large fault has occurred, which will also have an adverse impact on other components in the transmission system, thereby shortening the service life of the transmission system. Therefore, it is necessary to improve the sensitivity of fault detection so that the fault can be detected in a timely manner in the initial stage of the fault or when the fault is relatively minor, thereby improving the service life of the automotive transmission system.

[0055] First, when there is no fault in the transmission system under normal conditions, due to the coordinated operation between mechanical components, factors such as friction and meshing between components, the overall vibration situation is relatively similar. The vibration signals exhibit a certain degree of repeatability and regularity, and the vibration data differences are small. At the same time, since normal vibration is caused by rapidly cycling processes, such as gear rotation or shaft rotation, these processes have short time periods, so overall, it shows periodic changes with a relatively small period. When there is a fault in the transmission system, the vibration mode at the fault location will be different from normal vibration, resulting in an increase in vibration intensity, which in turn causes abnormalities in the vibration data and significant differences. At the same time, because the faulty components generate relatively irregular movements or impact loads, it will change the dynamic characteristics of the system, disturb the vibration period, and cause the vibration period to become longer. For example, non-uniform motion caused by bearing wear or gear damage, so overall, it shows periodic changes with a relatively large period.

[0056] Secondly, under normal conditions in the frequency domain, due to the influence of the meshing frequency and harmonic components, a relatively large amplitude will appear in the low-frequency region, but it is still mainly dominated by the meshing frequency. On both sides of the meshing frequency, the energy of the high-order harmonics gradually decays, that is, the amplitude change is relatively close to a parabolic distribution. Due to the existence of a fault, it will cause transient impacts, resulting in a modulation phenomenon. In the initial stage of the fault or in a relatively mild case, it is a mild and stable type of modulation, and the amplitude of the sidebands is generally relatively high and concentrated. As the fault gradually becomes more serious, it will evolve into an impact type of modulation, and the sidebands also show a parabolic shape, but there are obvious fluctuations in the medium and high-frequency data. Among them, the schematic diagram of the frequency-domain waveform diagram under normal conditions is as Figure 2 shown, the schematic diagram of the frequency-domain waveform diagram under a mild fault is as Figure 3 shown, and the schematic diagram of the frequency-domain waveform diagram under a severe fault is as Figure 4 shown.

[0057] (1) Take the vibration sequence of each period as the input of the autocorrelation function, and the output is a series of autocorrelation coefficients of the vibration sequences of each period. Each autocorrelation coefficient corresponds to a different lag amount. In the embodiments of the present application, the range of values of the lag amount for obtaining the autocorrelation coefficients is within [1, 100], and an autocorrelation function graph of the vibration sequence of this period is obtained based on the obtained autocorrelation coefficients. The schematic diagram of this autocorrelation function graph is as Figure 5 shown. As other embodiments of the present application, the implementer can set the range of the lag amount in the autocorrelation function graph according to the actual situation. In the autocorrelation function graph, the wave peaks with a peak value greater than the correlation threshold α are recorded as significant peaks. Preferably, in the embodiments of the present application, the value of α is set to 0.75. As other embodiments of the present application, the implementer can set the value of α according to the actual situation. Among them, the autocorrelation function is a well-known technology, and the specific process will not be elaborated.

[0058] Further, for each significant peak in the autocorrelation function graph, taking the lag corresponding to the peak value of the significant peak as the ordinate and the order in which the significant peak appears as the abscissa, the coordinate points corresponding to the significant peaks are obtained; the coordinate points corresponding to all the significant peaks in the autocorrelation function graph are linearly fitted by the least squares method, and the output fitted line is denoted as the lag fitted line. Among them, the least squares method is a well-known technology, and the specific process will not be elaborated.

[0059] (2) Taking the vibration sequence of each period as the input of the Fourier transform algorithm, and outputting the frequency-domain waveform diagram of the vibration sequence. Among them, the Fourier transform is a well-known technology, and the specific process will not be elaborated.

[0060] It should be noted that for the acquisition of the frequency-domain waveform diagram of the vibration sequence, this application only provides a time-frequency transformation method. There are many existing time-frequency transformation methods, and implementers can also use other time-frequency transformation algorithms to obtain the frequency-domain waveform diagram of the vibration sequence. This application does not make specific restrictions.

[0061] Calculating the mean value of the amplitudes of all frequencies in the frequency-domain waveform diagram as the frequency significance threshold; the frequencies with amplitudes greater than the frequency significance threshold are denoted as significant frequencies. Then, the entire frequency region in the frequency-domain waveform diagram is evenly divided into 3 segments to obtain 3 sub-intervals, which are denoted as the low-frequency region, the medium-frequency region, and the high-frequency region in ascending order of frequency. Further, arranging the amplitudes of all significant frequencies in the low-frequency region in ascending order of significant frequency to form a low-frequency amplitude sequence, and obtaining the first-order difference sequence of the low-frequency amplitude sequence. Among them, the first-order difference sequence is a well-known technology, and the specific process will not be elaborated.

[0062] (3) Based on the above analysis, calculating the vibration characteristic coefficients of each period to measure the time-frequency domain characteristics of the vibration signal of each period, specifically:

[0063] First, to analyze the periodicity of the vibration sequence of each period, calculating the product of the mean value of the peak values of all significant peaks in the autocorrelation function graph of each period and the coefficient of determination of the lag fitted line, which is denoted as the period characteristic value; among them, the calculation of the coefficient of determination is a well-known technology, and the specific process will not be elaborated;

[0064] Further, calculating the variance of all elements in the vibration sequence of each period, which is denoted as the first variance, to evaluate the fluctuation of the vibration sequence of each period;

[0065] Further, to analyze the amplitude difference between different frequency regions in the frequency domain of the vibration signal, in the frequency-domain waveform diagram of the vibration sequence of each period, calculating the average value of the variances of the amplitudes of all frequencies in the medium-frequency region and the high-frequency region, which is denoted as the first average value; at the same time, based on the analysis of the element changes in the first-order difference sequence of the low-frequency amplitude sequence, analyzing the low-frequency distribution characteristics of the low-frequency region in the frequency-domain waveform diagram, and the expression is: In the formula, lf i is the measurement of the low-frequency distribution characteristics in the low-frequency region of the frequency-domain waveform diagram of the i-th cycle, μ(L i ), σ(L i ) are respectively the mean and variance of all elements in the first-order difference sequence of the low-frequency amplitude sequence of the i-th cycle, and norm() is the normalization function.

[0066] Finally, calculate the vibration characteristic coefficient of each cycle, and the expression is:

[0067]

[0068] In the formula, A i is the vibration characteristic coefficient of the i-th cycle, R i is the cycle eigenvalue of the i-th cycle, σ(V i ) is the first variance of the i-th cycle, t i is the lag corresponding to the peak value of the first significant peak in the autocorrelation function diagram of the i-th cycle, mf i is the first mean value of the i-th cycle, lf i is the measurement of the low-frequency distribution characteristics in the low-frequency region of the frequency-domain waveform diagram of the i-th cycle, and exp() is the exponential function with the natural constant e as the base.

[0069] Under normal circumstances, the vibration data shows high periodic fluctuations and a small period, that is, R i is larger, t i is smaller; at the same time, the fluctuation of the vibration sequence is small, that is, σ(V i ) is smaller; secondly, the significant frequency amplitudes in the low-frequency region show a parabolic distribution, with the maximum in the middle and gradually decaying on both sides. Therefore, the mean value μ(L i ) of the first-order difference sequence is small, and the variance σ(L i ) is large, that is, lf i is small; at the same time, the amplitude fluctuation in the medium-high frequency region is small, that is, mf i is small, so exp(mf i -lf i ) is small; thus, under normal circumstances, the vibration characteristic coefficient A i is large. On the contrary, when a fault exists, the periodicity weakens, the period is small, the vibration data fluctuation increases, and the medium-high frequency amplitude fluctuation increases. Therefore, the vibration characteristic coefficient A i is small, and the more serious the fault, the smaller A i .

[0070] Step S3, constructing a total vibration sequence based on the vibration sequences of all cycles; constructing a vibration feature sequence based on the vibration feature coefficients of all cycles; and constructing a fault trend coefficient of the automobile transmission system based on the frequency of data fluctuations in the total vibration sequence and the data change trend in the vibration feature sequence.

[0071] The vibration characteristic coefficient determines the magnitude of the fault by analyzing the characteristics of the instantaneous vibration data. However, in the early stages of a fault or when the fault is relatively minor, the change in vibration data is not obvious, resulting in a smaller difference in the vibration characteristic coefficient. It is therefore difficult to monitor and diagnose the fault based solely on the vibration characteristic coefficient.

[0072] Under normal circumstances without any faults, the fluctuation of vibration data is small and has great regularity. At the same time, as analyzed above, the significant frequency amplitude fluctuation (sideband amplitude distribution) in the low-frequency area presents a parabolic distribution, and the medium and high frequency amplitudes are small and relatively stable. Therefore, the time series changes of vibration data are relatively stable as a whole, the frequency of vibration fluctuations is small, and there is no obvious trend in the fluctuations.

[0073] In the early stage of the fault, the vibration data will fluctuate to a certain extent, resulting in an increase in data fluctuations. At the same time, the sideband amplitude distribution will also change, and there will be high-frequency noise causing certain fluctuations in the mid- and high-frequency amplitudes. The data fluctuations caused by the above characteristics are relatively slight, and it is difficult to distinguish them from normal data fluctuations based on the vibration changes of a single cycle. However, due to the dual influence of the fault and normal data fluctuations, the vibration data will fluctuate more frequently in time series, and there will be a more obvious trend, that is, the fluctuations will become more and more intense. As the fault becomes more and more serious, the impact of the fault cause on the vibration data fluctuation becomes more and more dominant, which reduces the frequency of vibration data fluctuations, but the fluctuations are more intense, and the trend is more obvious.

[0074] (1) Arrange all elements of the vibration sequence of all cycles in ascending time order as a sequence as the total vibration sequence of the automobile transmission system, use the total vibration sequence as the input of the polynomial fitting algorithm, and output the equation of the fitting curve of the total vibration sequence. The polynomial fitting algorithm is a well-known technology, and the specific process will not be repeated here.

[0075] It should be noted that for the curve fitting of the total vibration sequence, this application only provides a curve fitting method. There are many existing curve fitting methods, and implementers can also use other curve fitting algorithms to perform curve fitting on the total vibration sequence. This application does not make specific restrictions.

[0076] (2) Denote the sequence formed by arranging the vibration characteristic coefficients of all cycles in ascending order of time as the vibration characteristic sequence. Use the vibration characteristic sequence as the input of the STL decomposition (Seasonal-Trend Decomposition using LOESS) algorithm to obtain the trend sequence of the vibration characteristic sequence. Use this trend sequence as the input of the least squares linear fitting algorithm, and the output is the fitting straight line equation of this trend sequence. Among them, both the STL decomposition algorithm and the least squares linear fitting algorithm are well-known technologies, and the specific process will not be elaborated here.

[0077] It should be noted that for the decomposition of the vibration characteristic sequence and the linear fitting of the trend sequence, this application only provides a time series decomposition method and a linear fitting method. There are many existing time series decomposition methods and linear fitting methods. Implementers can also use other time series decomposition algorithms and linear fitting algorithms to decompose the vibration characteristic sequence and linearly fit the trend sequence respectively. This application does not make specific restrictions.

[0078] (3) Based on the above analysis, calculate the fault trend coefficient, which is used to measure the change trend of the vibration of the transmission system. The expression is:

[0079] B = lg(X) × exp(-K), where B is the fault trend coefficient of the automotive transmission system, X is the number of data points with a derivative of 0 on the fitting curve of the total vibration sequence, K is the slope of the fitting straight line of the trend sequence, lg() is the logarithmic function with 10 as the true number, and exp() is the exponential function with the natural constant e as the base.

[0080] Since the frequency of vibration data fluctuations is relatively small both under normal conditions and in the case of relatively serious faults, the role of the logarithmic function is to reduce the influence of frequency measurement, and the role of the exponential function is to enhance the influence of trend measurement.

[0081] Under normal conditions, the frequency of vibration data fluctuations is low and there is no obvious trend, so X is small and |K| is also small, that is, lg(X) is small, and exp(-K) is close to 1, so the fault trend coefficient B is small. At the initial stage of the fault, the frequency of vibration data fluctuations increases and shows a certain trend as a whole, that is, X is large, K is negative and |K| increases, so lg(X) is large and exp(-K) also increases, so the fault trend coefficient B is large. As the fault gradually increases, the frequency of vibration data fluctuations also decreases, but the trend of vibration change is more obvious, that is, X decreases, K is negative and |K| is large, so lg(X) is relatively small and exp(-K) is large, but the exponential growth rate is faster than the logarithm, so the more serious the fault, the larger the fault trend coefficient B.

[0082] Step S4, determine the abnormal proportion adjustment value of the automotive transmission system based on the fault trend coefficient.

[0083] After obtaining the time-series monitoring data of the vibration data of the automotive transmission system, the abnormal points therein can be detected through an anomaly detection algorithm. However, in the traditional isolation forest algorithm, when performing anomaly point detection, each parameter set therein is fixed. In this way, the set abnormal proportion is often small, which can reduce the false alarm rate of faults. This leads to a better fault detection effect when the fault characteristics are obvious. However, in the initial stage of the fault or when the fault is relatively minor, the difference between the data is not obvious enough, making it difficult to detect abnormal points. If the abnormal proportion is set larger, although it can improve the sensitivity of anomaly detection and have a better fault recognition ability in the initial stage of the fault or when the fault is relatively minor, it will increase the false alarm rate of faults. Therefore, the selection of the abnormal proportion affects the sensitivity and accuracy of the isolation forest algorithm for anomaly detection, and is also the key to restricting the application of this algorithm to the fault diagnosis of the automotive transmission system. Therefore, it is necessary to adaptively adjust the abnormal proportion according to the vibration situation of the automotive transmission system to improve the sensitivity and accuracy of anomaly detection. The expression is:

[0084] In the formula, W is the abnormal proportion adjustment value of the automotive transmission system, B is the fault trend coefficient of the automotive transmission system, and arctan() is the arctangent function.

[0085] When the fault is more serious, the fault trend coefficient B is larger, and the abnormality of the vibration data is more obvious. At this time, the corresponding abnormal proportion is smaller, which can reduce the false alarm rate; on the contrary, in the initial stage of the fault or when the fault is relatively minor, the abnormal change of the vibration data is more minor. At this time, the corresponding abnormal proportion is larger, so as to improve the sensitivity of anomaly recognition and reduce the missed alarm rate.

[0086] Step S5, based on the abnormal proportion adjustment value, combine with the isolation forest algorithm to monitor the operation faults of the automotive transmission system.

[0087] Take the vibration sequences of all collected cycles as the input of the isolation forest algorithm. The number of samples extracted each time is 256, and the number of isolation trees in the forest is 100. Take the calculated abnormal proportion adjustment value W above as the abnormal proportion in the isolation forest algorithm, and the output is the abnormal score of each cycle of vibration sequence. Take the abnormal scores of the vibration sequences of all cycles as the input of Otsu's method, and the output is the segmentation threshold. Among them, the isolation forest algorithm and Otsu's method are both well-known technologies, and the specific process will not be elaborated.

[0088] It should be noted that for the setting of the number of samples extracted each time and the number of isolation trees, only one setting method is provided in the embodiments of the present application. Implementers can set the number of samples extracted each time and the number of isolation trees according to the actual situation, and the present application does not make specific restrictions.

[0089] The periods with abnormal scores of the vibration sequence greater than the segmentation threshold are taken as abnormal periods. All the moments corresponding to all the abnormal periods are arranged in ascending order of time to form an abnormal sequence. The first-order difference sequence of this abnormal sequence is obtained, and the Hurst exponent of this first-order difference sequence is calculated. When the Hurst exponent is greater than the fault threshold, it indicates that there is an operating fault in the current automotive transmission system, and relevant maintenance personnel are required to repair it; when the Hurst exponent is less than or equal to the fault threshold, it indicates that the current automotive transmission system is operating normally. Preferably, in the embodiments of the present application, the fault threshold is set to 0.5. As other embodiments of the present application, implementers can set the fault threshold according to the actual situation. Among them, the Hurst exponent is a well-known technology, and the specific process will not be elaborated.

[0090] Based on the same inventive concept as the above method, the embodiments of the present application also provide an operating fault monitoring system applied to an automotive transmission system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above methods for the operating fault monitoring method applied to the automotive transmission system.

[0091] In summary, the embodiments of the present application provide an operating fault monitoring method applied to an automotive transmission system. According to the time-frequency domain characteristics of the instantaneous vibration data of each cycle during the operation of the automotive transmission system, the vibration characteristic coefficient is calculated, which helps to improve the accuracy of identifying the difference between normal and abnormal states; then, combined with the time-series change trend of the vibration data and the vibration characteristic coefficient, the fault trend coefficient is calculated, which helps to improve the accuracy and sensitivity of subsequent fault diagnosis; finally, the abnormal ratio is adaptively adjusted according to the fault trend coefficient, balancing the accuracy and sensitivity of the isolation forest anomaly detection. In this way, by combining the time-frequency domain characteristics and the time-series change trend of the vibration data, the sensitivity of anomaly detection in the initial stage of the fault or in the case of minor faults is improved, and it can respond flexibly according to the actual vibration situation, improving the adaptability and robustness of anomaly detection, ensuring the dynamic balance of the false alarm rate and the missed alarm rate of the transmission system fault detection, improving the quality of the automotive transmission system fault detection, and thus realizing the high-precision and high-sensitivity operating fault monitoring of the automotive transmission system, ensuring the safety of the automotive transmission system and extending the service life to a certain extent.

[0092] It should be noted that: the above-mentioned sequence of embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. And the above-mentioned specific embodiments of the present application have been described. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] The embodiments in the present application are all described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other, and the key points of each embodiment are the differences from other embodiments.

[0094] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for monitoring operating faults in an automobile transmission system, characterized in that: The method comprises the following steps: Collect all vibration signals in each cycle when the vehicle transmission system is running, and construct the vibration sequence of each cycle; Based on the periodic variation characteristics of the elements in the vibration sequence of each period, the fluctuation of the vibration sequence, and the amplitude difference between different frequency regions of the vibration sequence in the frequency domain, the vibration characteristic coefficient of each period is determined; The total vibration sequence is constructed based on the vibration sequence of all cycles; the vibration characteristic sequence is constructed based on the vibration characteristic coefficients of all cycles; the fault trend coefficient of the automobile transmission system is constructed based on the frequency of data fluctuations in the total vibration sequence and the data change trend in the vibration characteristic sequence; Determining an abnormal ratio adjustment value of a vehicle transmission system based on the fault trend coefficient; Based on the abnormal proportion adjustment value, the automobile transmission system operation fault monitoring is performed in combination with the isolation forest algorithm.

2. The operating fault monitoring method for an automobile transmission system according to claim 1, characterized in that: The process of obtaining the vibration characteristic coefficient of each period is as follows: The autocorrelation coefficient of the vibration sequence of each period is obtained through the autocorrelation function, and the autocorrelation function diagram of each period of the vibration sequence is constructed; based on the peak value distribution characteristics in the autocorrelation function diagram, the period characteristic value of the autocorrelation function diagram of each period is constructed; Calculate the variance of all elements in the vibration sequence of each period, recorded as the first variance; A frequency domain waveform diagram of each periodic vibration sequence is obtained by a time-frequency conversion algorithm. In each frequency domain waveform diagram, all frequencies are divided into a low-frequency region, a medium-frequency region, and a high-frequency region; an average value of the variance of the amplitudes of all frequencies in the medium-frequency region and the high-frequency region is calculated, and recorded as a first average value; Recording frequencies in the frequency domain waveform graph whose amplitudes are greater than a preset frequency significance threshold as significant frequencies, and recording a sequence consisting of the amplitudes of all significant frequencies in the low-frequency region as a low-frequency amplitude sequence; Obtaining a low-frequency distribution feature measure of the low-frequency region based on changes in elements in a first-order difference sequence of a low-frequency amplitude sequence; The vibration characteristic coefficient of each period is constructed based on the period characteristic value, the first variance, and the difference between the first average value and the low-frequency distribution characteristic metric.

3. The operating fault monitoring method for an automobile transmission system according to claim 2, characterized in that: The process of obtaining the periodic characteristic value is as follows: The peaks in the autocorrelation function graph of each period whose peak values ​​are greater than the preset correlation threshold are recorded as significant peaks; The hysteresis corresponding to each significant peak value is used as the ordinate, and the order of appearance of each significant peak is used as the abscissa to obtain the coordinate point corresponding to each significant peak; the coordinate points corresponding to all significant peaks in the autocorrelation function diagram are fitted by a linear fitting algorithm, and the obtained fitting straight line is recorded as the hysteresis fitting straight line; and the determination coefficient of the hysteresis fitting straight line is calculated; The product of the mean value of all significant peak-to-peak values ​​in the autocorrelation function diagram of each period and the determination coefficient is taken as the period characteristic value of the autocorrelation function diagram of each period.

4. The operating fault monitoring method for an automobile transmission system according to claim 2, characterized in that: The expression of the low-frequency distribution feature measurement is: In the formula, lf i is the low-frequency distribution characteristic measure of the low-frequency region in the frequency domain waveform of the ith period, μ(L i ),σ(L i ) are respectively the mean and variance of all elements in the first-order difference sequence of the low-frequency amplitude sequence of the ith period, and norm() is the normalization function.

5. The operating fault monitoring method for automobile transmission system according to claim 3, characterized in that: The expression of the vibration characteristic coefficient is: In the formula, A i is the vibration characteristic coefficient of the ith period, R i is the periodic characteristic value of the ith period, σ(V i ) is the first variance of the ith period, t i is the lag corresponding to the first significant peak in the autocorrelation function diagram of the ith period, mf i is the first average value of the ith period, lf i is the low-frequency distribution characteristic measure of the ith period, and exp() is an exponential function with the natural constant e as the base.

6. The operating fault monitoring method for an automobile transmission system according to claim 1, characterized in that: The process of obtaining the failure trend coefficient of the automobile transmission system is as follows: The total vibration sequence is used as the input of the polynomial fitting algorithm, and the output is the fitting curve of the total vibration sequence; Using the time series decomposition algorithm, the trend sequence of the vibration characteristic sequence is obtained; A fault trend coefficient of the automobile transmission system is constructed based on changes in the fitting curve and the fitting straight line of the trend sequence.

7. The operating fault monitoring method for automobile transmission system according to claim 6, characterized in that: The expression of the fault tendency coefficient is: B=lg(X)×exp(-K), where B is the failure trend coefficient of the automobile transmission system, X is the number of data points whose derivative is 0 on the fitting curve, K is the slope of the fitting straight line of the trend sequence, lg() is a logarithmic function with 10 as a real number, and exp() is an exponential function with the natural constant e as the base.

8. The operating fault monitoring method for automobile transmission system according to claim 1, characterized in that: The expression of the abnormal ratio adjustment value of the automobile transmission system is: Where W is the abnormal ratio adjustment value of the automobile transmission system, B is the failure trend coefficient of the automobile transmission system, and arctan() is the inverse tangent function.

9. The operating fault monitoring method for automobile transmission system according to claim 1, characterized in that: The automobile transmission system operation fault monitoring is performed based on the abnormal proportion adjustment value and in combination with the isolation forest algorithm, specifically: The vibration sequences of all cycles are used as input of the isolation forest algorithm, the abnormal proportion adjustment value is used as the abnormal proportion in the isolation forest algorithm, and the abnormal score of each cycle vibration sequence is output; the segmentation threshold of the abnormal score of the vibration sequence of all cycles is obtained; The period in which the abnormal score of the vibration sequence is greater than the segmentation threshold is regarded as an abnormal period; the sequence consisting of all moments corresponding to all abnormal periods is obtained and recorded as an abnormal sequence, and the Hurst index of the first-order difference sequence of the abnormal sequence is calculated. If the Hurst index is greater than the preset fault threshold, the current vehicle transmission system has an operating fault; otherwise, the current vehicle transmission system operates normally.

10. An operation fault monitoring system for an automobile transmission system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images