Fault part determination method, device and equipment and readable storage medium

Abstract

The invention discloses a fault part determination method, device and equipment and a readable storage medium. The fault part determination method comprises the following steps: determining a fault frequency band; wavelet packet decomposition is carried out on the vibration signal to obtain a plurality of sub-band signals, and the sub-band signals in a fault frequency band form a fault vibration signal; performing empirical mode decomposition on the fault vibration signal to obtain a plurality of intrinsic mode functions, and calculating a correlation coefficient between each intrinsic mode function and the fault vibration signal; aiming at the intrinsic mode function of which the correlation coefficient is greater than a preset coefficient, carrying out Hilbert transform on the intrinsic mode function, solving an envelope signal of the intrinsic mode function, and carrying out Fourier transform on the envelope signal to obtain a second frequency domain signal; and the difference value between the frequency corresponding to the maximum amplitude signal in the second frequency domain signal and the characteristic frequency of each part of the power assembly is calculated, and the part with the difference value smaller than a preset difference value serves as a fault part. According to the invention, the fault part can be efficiently and accurately locked.

Description

Technical Field

[0001] This application relates to the field of vehicle fault diagnosis technology, and in particular to a method, apparatus, device and readable storage medium for determining faulty components. Background Technology

[0002] The powertrain of a vehicle mainly consists of an engine / motor and a transmission system (including clutch, gearbox, drive shaft, differential and half shaft, etc.). The highly integrated nature of the powertrain means that abnormal vibration signals of the internal components cannot be collected at close range. Vibration signals can only be collected from the outside of the powertrain and then compared with the vibration signals under normal conditions to determine whether there are abnormal noises in the powertrain.

[0003] After determining that there is an abnormal noise in the powertrain, the usual method is to replace normal parts to verify and pinpoint the faulty part. However, the powertrain has many internal parts, and it is inefficient to repeatedly replace normal parts to verify and pinpoint the faulty part.

[0004] In summary, currently, when abnormal noise occurs in the powertrain, it is impossible to efficiently pinpoint the faulty component causing the noise. Summary of the Invention

[0005] This application provides a method, apparatus, device, and readable storage medium for determining faulty components, aiming to solve the current technical problem of being unable to efficiently locate the faulty components that generate abnormal noise after abnormal noise occurs in the powertrain.

[0006] In a first aspect, embodiments of this application provide a method for determining faulty components, the method comprising: Vibration signals were collected during the periods when abnormal noises occurred in the powertrain; The vibration signal is converted into a first frequency domain signal by Fourier transform, and the fault frequency range is determined based on the amplitude of the first frequency domain signal. Wavelet packet decomposition of the vibration signal yields multiple sub-band signals, each of which is located in a different frequency band. The sub-band signals located in the fault frequency band are combined to form the fault vibration signal. The fault vibration signal is subjected to empirical mode decomposition to obtain multiple intrinsic mode functions, and the correlation coefficient between each intrinsic mode function and the fault vibration signal is calculated. For intrinsic mode functions with correlation coefficients greater than preset coefficients, Hilbert transform is performed on the intrinsic mode functions, and the envelope signal of the intrinsic mode functions is solved. Fourier transform is then performed on the envelope signal to obtain the second frequency domain signal. Calculate the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the characteristic frequency of each component of the powertrain, and identify the component whose difference is less than a preset difference as the faulty component.

[0007] Optionally, determining the fault frequency band based on the amplitude of the first frequency domain signal includes: The first frequency domain signal is divided into multiple frequency intervals; For each frequency range, if the average signal amplitude within the frequency range is greater than the preset amplitude, then the frequency range is designated as the fault frequency segment.

[0008] Optionally, the correlation coefficient is the Pearson correlation coefficient, and calculating the correlation coefficient between each intrinsic mode function and the fault vibration signal includes: The Pearson correlation coefficient between each intrinsic mode function and the fault vibration signal is calculated using Formula 1, which is: ; Where C is the Pearson correlation coefficient, and n is the total number of signal sampling points. This represents the value of each intrinsic mode function at the i-th sampling point. This represents the mean of all sampled points for each intrinsic mode function. Let be the value of the fault vibration signal at the i-th sampling point. This is the mean value of all sampling points of the fault vibration signal.

[0009] Optionally, performing a Hilbert transform on the intrinsic mode functions and solving for the envelope signal of the intrinsic mode functions includes: By performing a Hilbert transform on the intrinsic mode functions, the orthogonal components of the intrinsic mode functions are obtained; The eigenmode functions and orthogonal components are combined to obtain the analytic signal; The envelope signal of the intrinsic mode function is obtained by taking the modulus of the analytic signal.

[0010] Optionally, the characteristic frequencies of the components include first-order characteristic frequencies, second-order characteristic frequencies, and third-order characteristic frequencies. The step of calculating the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the characteristic frequency of each component in the powertrain, and identifying components with a difference less than a preset difference as faulty components, includes: For each component, calculate the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the first-order characteristic frequency, second-order characteristic frequency and third-order characteristic frequency of the component; If any difference is less than the preset difference, the component will be considered a faulty component.

[0011] Optionally, before calculating the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the first-order characteristic frequency, second-order characteristic frequency, and third-order characteristic frequency of the component for each component, the following steps are included: For each component, the first-order characteristic frequency of the component is calculated based on the component's fundamental frequency and rotational speed, where the rotational speed is determined based on the vehicle speed at the time of signal acquisition. The second-order and third-order characteristic frequencies of the components are calculated based on their first-order characteristic frequencies.

[0012] Optionally, the calculation of the first-order characteristic frequency of the component based on its fundamental frequency and rotational speed includes: The first-order characteristic frequency of the component is calculated using Formula 2 based on the component's fundamental frequency and rotational speed. Formula 2 is as follows: The first-order characteristic frequency of a component = the fundamental frequency of the component × (rotation speed / 60).

[0013] Optionally, the period during which abnormal noise occurs in the powertrain can be determined based on the sound signals collected by the microphone.

[0014] Optionally, the preset amplitude is a preset multiple of the average amplitude of the vibration signal during the period when the powertrain is free of abnormal noise.

[0015] Secondly, embodiments of this application provide a faulty component determination device, the faulty component determination device comprising: The acquisition module is used to collect vibration signals during periods when abnormal noises occur in the powertrain; The first conversion module is used to convert the vibration signal into a first frequency domain signal by performing a Fourier transform, and to determine the fault frequency range based on the amplitude of the first frequency domain signal. The decomposition module is used to perform wavelet packet decomposition on the vibration signal to obtain multiple sub-band signals, where each sub-band signal is in a different frequency band, and the sub-band signals in the fault frequency band are combined to form the fault vibration signal. The calculation module is used to perform empirical mode decomposition on the fault vibration signal to obtain multiple intrinsic mode functions, and calculate the correlation coefficient between each intrinsic mode function and the fault vibration signal; The second conversion module is used to perform Hilbert transform on the intrinsic mode functions with correlation coefficients greater than preset coefficients, solve for the envelope signal of the intrinsic mode functions, and perform Fourier transform on the envelope signal to obtain the second frequency domain signal. The comparison module is used to calculate the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the characteristic frequency of each component of the powertrain, and to identify the component whose difference is less than a preset difference as the faulty component.

[0016] Optionally, the determination of the fault frequency band based on the amplitude of the first frequency domain signal is used for: The first frequency domain signal is divided into multiple frequency intervals; For each frequency range, if the average signal amplitude within the frequency range is greater than the preset amplitude, then the frequency range is designated as the fault frequency segment.

[0017] Optionally, the correlation coefficient is the Pearson correlation coefficient, and the calculation of the correlation coefficient between each intrinsic mode function and the fault vibration signal is used for: The Pearson correlation coefficient between each intrinsic mode function and the fault vibration signal is calculated using Formula 1, which is: ; Where C is the Pearson correlation coefficient, and n is the total number of signal sampling points. This represents the value of each intrinsic mode function at the i-th sampling point. This represents the mean of all sampled points for each intrinsic mode function. Let be the value of the fault vibration signal at the i-th sampling point. This is the mean value of all sampling points of the fault vibration signal.

[0018] Optionally, the Hilbert transform of the intrinsic mode functions and the solution of the envelope signal of the intrinsic mode functions are used for: By performing a Hilbert transform on the intrinsic mode functions, the orthogonal components of the intrinsic mode functions are obtained; The eigenmode functions and orthogonal components are combined to obtain the analytic signal; The envelope signal of the intrinsic mode function is obtained by taking the modulus of the analytic signal.

[0019] Thirdly, embodiments of this application provide a faulty component determination device, which includes a processor, a memory, and a faulty component determination program stored in the memory and executable by the processor, wherein when the faulty component determination program is executed by the processor, it implements the steps of the faulty component determination method as described above.

[0020] Fourthly, embodiments of this application provide a readable storage medium storing a faulty component determination program, wherein when the faulty component determination program is executed by a processor, it implements the steps of the faulty component determination method as described above.

[0021] The beneficial effects of the technical solutions provided in this application include: In this embodiment, vibration signals during the period when abnormal noise occurs in the powertrain are collected; the vibration signals are converted into a first frequency domain signal by Fourier transform, and the fault frequency band is determined based on the amplitude of the first frequency domain signal; the vibration signals are decomposed into multiple sub-band signals by wavelet packet decomposition, wherein each sub-band signal is in a different frequency band, and the sub-band signals in the fault frequency band are combined to form the fault vibration signal; the fault vibration signal is decomposed into empirical mode functions to obtain multiple intrinsic mode functions, and the correlation coefficient between each intrinsic mode function and the fault vibration signal is calculated; for intrinsic mode functions with correlation coefficients greater than a preset coefficient, Hilbert transform is performed on the intrinsic mode functions, and the envelope signal of the intrinsic mode functions is solved, and Fourier transform is performed on the envelope signal to obtain a second frequency domain signal; the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the characteristic frequency of each component of the powertrain is calculated, and the component with a difference less than a preset difference is identified as the faulty component. In this embodiment, for the vibration signal during the period when the powertrain is making abnormal noise, it is first converted from a time-domain signal to a frequency-domain signal using Fourier transform. Based on the frequency-domain signal, including the frequency-amplitude correspondence spectrum, the fault frequency range with larger amplitude is selected. Further, wavelet packet decomposition, empirical mode decomposition, and Hilbert transform are used to improve the signal-to-noise ratio of the signal within the fault frequency range and eliminate interference from signals outside the fault frequency range. Then, for the converted second frequency domain signal, the frequency corresponding to the fault signal is determined by the maximum amplitude. The frequency corresponding to the fault signal is compared with the inherent characteristic frequency of each component. The characteristic frequency of the component is determined by its inherent properties. By comparing the difference between the two, if the difference is less than a preset difference, it is indicated that the component is faulty. Thus, by analyzing and processing the vibration signal during the period when the powertrain is making abnormal noise, the faulty component can be located efficiently and accurately without the need for repeated manual replacement of normal components for verification, which greatly improves the efficiency of locating the faulty component when the powertrain is making abnormal noise. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating an embodiment of the method for determining faulty components according to this application; Figure 2 This is a schematic diagram of vibration signals from one embodiment of the method for determining faulty components according to this application. Figure 3 This is a schematic diagram of the first frequency domain signal of an embodiment of the method for determining faulty components in this application; Figure 4 This is a schematic diagram of wavelet packet decomposition of an embodiment of the faulty component determination method of this application; Figure 5 This is a schematic diagram of empirical mode decomposition of an embodiment of the method for determining faulty components in this application; Figure 6 This is a schematic diagram of the envelope signal of one embodiment of the faulty component determination method of this application; Figure 7 This is a schematic diagram of the second frequency domain signal of an embodiment of the faulty component determination method of this application; Figure 8 This is a functional module diagram of an embodiment of the faulty component identification device of this application; Figure 9 This is a schematic diagram of the hardware structure of the device for determining faulty components involved in the embodiments of this application. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0025] In a first aspect, embodiments of this application provide a method for determining faulty components.

[0026] In one embodiment, reference is made to Figure 1 , Figure 1 This is a flowchart illustrating an embodiment of the method for determining faulty components according to this application, as shown below. Figure 1 As shown, the method for determining faulty components includes: Step S10: Collect vibration signals during the period when abnormal noise occurs in the powertrain.

[0027] In this embodiment, the vehicle's powertrain is highly integrated, making it impossible to collect abnormal vibration signals from internal components at close range. Therefore, high-sensitivity vibration sensors are placed on the outer surface of the powertrain to collect vibration signals during operation in real time (the sampling frequency is set to 20 kHz or higher). The collected vibration signals during the period when abnormal noise occurs in the powertrain are time-domain signals, as referenced... Figure 2 , Figure 2 This is a schematic diagram of vibration signals from one embodiment of the method for determining faulty components according to this application, as shown below. Figure 2As shown, the horizontal axis represents time in seconds (s), and the vertical axis represents the real part of the vibration signal, forming a curve that changes over time. The green line represents the vibration response of a "normal powertrain," and the red line represents the vibration response of an "abnormal powertrain." Peaks in the red area are labeled as "impact signals," representing fault characteristics. The placement of vibration sensors should comprehensively consider the locations of key components of the powertrain (such as the engine block, transmission housing, and driveshaft connections) to fully capture fault vibration signals. Microphones can be placed near the powertrain to collect sound signals. By analyzing the time and frequency domain characteristics of the sound signals, the time period of the abnormal noise can be determined, and the corresponding vibration signal can be identified.

[0028] Step S20: Perform Fourier transform on the vibration signal to convert it into a first frequency domain signal, and determine the fault frequency band based on the amplitude of the first frequency domain signal.

[0029] In this embodiment, the vibration signal collected during the period when the powertrain emitted abnormal noise is a time-domain signal. This signal is converted to a frequency-domain signal using a Fast Fourier Transform (FFT) to obtain the signal's spectrum. (Refer to...) Figure 3 , Figure 3 This is a schematic diagram of the first frequency domain signal of an embodiment of the faulty component determination method of this application, as shown below. Figure 3 As shown, the spectrum diagram can display the amplitude distribution of a signal at different frequencies. Figure 3 The horizontal axis represents frequency, measured in Hertz (Hz), which is the number of vibration cycles per second, ranging from 0 to 10000 Hz. It represents the analysis of vibration components from low to mid-high frequencies. Each frequency point corresponds to a sine wave component. In mechanical systems, common fault characteristic frequencies (such as rotational speed, bearing failure frequency, and gear meshing frequency) appear in specific frequency bands. The vertical axis represents acceleration amplitude, measured in g (the unit of gravitational acceleration), where 1 g ≈ 9.8 m / s². This represents the intensity of vibration energy at that frequency; a larger value indicates a stronger frequency component, which may indicate the presence of resonance or a fault source. By dividing the spectrum into multiple frequency segments (e.g., every 10 Hz), the average signal amplitude within each frequency segment is calculated and compared with a preset amplitude (e.g., set to 1.5 times the average vibration signal amplitude during periods when the powertrain has no abnormal noise). Abnormal vibration signals typically exhibit a significant increase in amplitude within a specific frequency range. If the average amplitude within a certain frequency range exceeds a preset amplitude, that frequency range is identified as the fault frequency range. Figure 3 As shown, four fault frequency bands can be identified as indicated by the red box. Spectrum analysis can quickly pinpoint the fault frequency range, providing a precise target frequency range for subsequent analysis, avoiding blind analysis, and improving the efficiency of identifying faulty powertrain components.

[0030] Step S30: Wavelet packet decomposition is performed on the vibration signal to obtain multiple sub-band signals, where each sub-band signal is in a different frequency band. The sub-band signals in the fault frequency band are combined to form the fault vibration signal.

[0031] In this embodiment, refer to Figure 4 , Figure 4 This is a schematic diagram of wavelet packet decomposition of an embodiment of the faulty component determination method of this application, as shown below. Figure 4 As shown, wavelet packet decomposition is performed on the vibration signal. By selecting an appropriate mother wavelet (such as the db4 wavelet) and the number of decomposition levels (such as 4-6 levels), the signal is decomposed into 16 sub-band signals, each corresponding to a different frequency band. Then, the sub-band signals in the fault frequency band are combined to form the fault vibration signal. Wavelet packet decomposition provides finer resolution in the time-frequency plane and can better separate different frequency components, especially showing good analytical capabilities for non-stationary signals (such as vibration signals). Through wavelet packet decomposition and signal recombination, noise signals outside the fault frequency band can be effectively removed, improving the signal-to-noise ratio of the fault signal and providing a clearer signal for subsequent fault feature extraction, avoiding the problem of difficulty in distinguishing fault signals from noise in traditional frequency domain analysis.

[0032] Step S40: Perform empirical mode decomposition on the fault vibration signal to obtain multiple intrinsic mode functions, and calculate the correlation coefficient between each intrinsic mode function and the fault vibration signal.

[0033] In this embodiment, refer to Figure 5 , Figure 5 This is a schematic diagram of empirical mode decomposition of an embodiment of the faulty component determination method of this application, as shown below. Figure 5 As shown, the fault vibration signal is subjected to Empirical Mode Decomposition (EMD) to obtain 14 intrinsic mode functions (IMFs) IMF1-IMF14. EMD is an adaptive signal decomposition method that decomposes the signal into physically meaningful IMFs, each representing an oscillation component of the signal at different frequency scales. Then, the correlation coefficient between each IMF and the fault vibration signal is calculated, using the Pearson correlation coefficient formula. Referring to Table 1, the correlation coefficients between IMFs IMF1-IMF14 and the fault vibration signal are calculated. The IMF components highly correlated with the fault signal are selected based on the correlation coefficients; these components contain the main fault information. The adaptive nature of EMD allows it to decompose the signal according to its characteristics, and by adaptively extracting fault features, the accuracy of fault feature extraction can be improved.

[0034] Table 1

[0035] Step S50: For the intrinsic mode functions with correlation coefficients greater than preset coefficients, perform Hilbert transform on the intrinsic mode functions, solve for the envelope signal of the intrinsic mode functions, and perform Fourier transform on the envelope signal to obtain the second frequency domain signal.

[0036] In this embodiment, for the intrinsic mode functions (IMFs) with a correlation coefficient greater than a preset coefficient (e.g., 0.6), and since only IMF1 in Table 1 has a correlation coefficient greater than 0.6, a Hilbert transform is performed on IMF1. The Hilbert transform converts the real signal into an analytic signal. Then, the modulus of the analytic signal is taken to obtain the envelope signal. The obtained envelope signal can be referred to... Figure 6 As shown, Figure 6 This is a schematic diagram of the envelope signal of one embodiment of the faulty component determination method of this application. The envelope signal reflects the amplitude modulation characteristics of the signal and can highlight the instantaneous amplitude changes of the signal. Finally, a Fourier transform is performed on the envelope signal to obtain a second frequency domain signal, which can be referred to... Figure 7 As shown, Figure 7 This is a schematic diagram of the second frequency domain signal according to an embodiment of the fault component determination method of this application. By using Hilbert transform and envelope analysis, the instantaneous features of the signal can be extracted more accurately, effectively removing noise interference, because the envelope signal mainly reflects the amplitude change of the signal, and the amplitude change is closely related to the fault source. By further improving the identification accuracy of the fault frequency, the fault characteristics become more obvious.

[0037] Step S60: Calculate the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the characteristic frequency of each component of the powertrain, and identify the component whose difference is less than the preset difference as the faulty component.

[0038] In this embodiment, the characteristic frequencies of each component in the powertrain are related to its inherent properties (such as the number of bearing balls, the number of gear teeth, etc.) and operating state (speed). By matching these frequencies with the fault frequencies, the faulty component can be identified. (Continue referring to...) Figure 7 The frequency corresponding to the largest amplitude signal in the second frequency domain signal can be used as the fault frequency. Then, the difference between the fault frequency and the inherent characteristic frequency of each component of the powertrain is compared. The component with a difference less than the preset difference is identified as the fault component. Thus, the fault component can be located efficiently and accurately by analyzing and processing the vibration signal during the period when the powertrain makes abnormal noise. There is no need to manually replace normal components repeatedly for verification, which greatly improves the efficiency of locating fault components when the powertrain makes abnormal noise.

[0039] In this embodiment, the vehicle's powertrain is highly integrated, making it impossible to collect abnormal vibration signals from internal components at close range. By deploying high-sensitivity vibration sensors on the outer surface of the powertrain, vibration signals during its operation are collected in real time. The collected vibration signals during the periods when abnormal noise occurs are time-domain signals, which are converted to frequency-domain signals using a Fast Fourier Transform (FFT) to obtain a spectrum. The spectrum shows the amplitude distribution of the signal at different frequencies. Spectrum analysis can quickly pinpoint the fault frequency range, providing a precise target frequency range for subsequent analysis and avoiding blind analysis. By performing wavelet packet decomposition and signal recombination on the vibration signal, noise signals outside the fault frequency range can be effectively removed, improving the signal-to-noise ratio of the fault signal. To provide clearer signals for subsequent fault feature extraction, empirical mode decomposition is performed on the fault vibration signal. IMF components that are highly correlated with the fault signal are screened out by correlation coefficient, and fault features are adaptively extracted to improve the accuracy of fault feature extraction. Hilbert transform and envelope analysis are used to extract the instantaneous features of the signal more accurately and effectively remove noise interference. The characteristic frequencies of each component in the powertrain are related to their inherent properties and operating state (speed). By matching them with the fault frequency, the faulty component can be identified. Therefore, the faulty component can be efficiently and accurately located by analyzing and processing the vibration signal during the period when the powertrain makes abnormal noise, without the need for manual and repeated replacement of normal components for verification. This greatly improves the efficiency of locating the faulty component when the powertrain makes abnormal noise.

[0040] Furthermore, in one embodiment, determining the fault frequency band based on the amplitude of the first frequency domain signal includes: The first frequency domain signal is divided into multiple frequency intervals; For each frequency range, if the average signal amplitude within the frequency range is greater than the preset amplitude, then the frequency range is designated as the fault frequency segment.

[0041] In this embodiment, for example, the first frequency domain signal is divided into multiple frequency segments at 10Hz intervals, the average amplitude of each frequency segment is calculated, and compared with a preset amplitude (e.g., 1.5 times the average amplitude of the vibration signal during a period when the powertrain has no abnormal noise). If the average amplitude of a certain frequency segment is greater than the preset amplitude, then that frequency segment is determined as the fault frequency segment. Figure 3 As shown, four fault frequency ranges can be identified, as indicated by the red box. When abnormal noise occurs in the powertrain, the fault signal usually shows a significant increase in amplitude within a specific frequency range. Spectrum analysis can quickly pinpoint the fault frequency range, providing a precise target frequency range for subsequent analysis. By setting reasonable preset amplitude values, fault signals and normal signals can be effectively distinguished.

[0042] Further, in one embodiment, the correlation coefficient is the Pearson correlation coefficient, and calculating the correlation coefficient between each intrinsic mode function and the fault vibration signal includes: The Pearson correlation coefficient between each intrinsic mode function and the fault vibration signal is calculated using Formula 1, which is: ; Where C is the Pearson correlation coefficient, and n is the total number of signal sampling points. This represents the value of each intrinsic mode function at the i-th sampling point. This represents the mean of all sampled points for each intrinsic mode function. Let be the value of the fault vibration signal at the i-th sampling point. This is the mean value of all sampling points of the fault vibration signal.

[0043] In this embodiment, the characteristic components of the fault signal at different frequency scales are strongly correlated with the original fault signal. By filtering using correlation coefficients, the main components containing fault information can be accurately extracted. The Pearson correlation coefficient between each intrinsic mode function (IMF1-IMF14) and the fault vibration signal is calculated using Formula 1. The calculated correlation coefficients of IMF1-IMF14 are shown in Table 1, and the IMF components highly correlated with the fault signal are selected. The Pearson correlation coefficient accurately reflects the degree of linear correlation between two signals, with a value range of [-1, 1]. The closer the value is to 1, the stronger the correlation. Based on the Pearson correlation coefficient, the accuracy of fault feature extraction can be improved.

[0044] Further, in one embodiment, performing a Hilbert transform on the intrinsic mode functions and solving for the envelope signal of the intrinsic mode functions includes: By performing a Hilbert transform on the intrinsic mode functions, the orthogonal components of the intrinsic mode functions are obtained; The eigenmode functions and orthogonal components are combined to obtain the analytic signal; The envelope signal of the intrinsic mode function is obtained by taking the modulus of the analytic signal.

[0045] In this embodiment, a Hilbert transform is performed on each IMF to obtain its orthogonal components. Then, the IMF and its orthogonal components are combined to form an analytic signal, the magnitude of which is the envelope signal. The Hilbert transform converts a real signal into an analytic signal. The magnitude of the analytic signal reflects the instantaneous amplitude of the signal, highlighting its amplitude modulation characteristics, which are closely related to the fault source. This allows for more accurate extraction of the signal's instantaneous features, effective removal of noise interference, and improved accuracy in fault frequency identification.

[0046] Furthermore, in one embodiment, the characteristic frequencies of the component include first-order characteristic frequencies, second-order characteristic frequencies, and third-order characteristic frequencies, and step S60 includes: For each component, calculate the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the first-order characteristic frequency, second-order characteristic frequency and third-order characteristic frequency of the component; If any difference is less than the preset difference, the component will be considered a faulty component.

[0047] In this embodiment, the vibration signals of rotating machinery are typically described by order. Fault signals often appear at integer multiples of harmonics (such as first, second, or third order) rather than fixed frequencies. Therefore, for each component, the frequency corresponding to the largest amplitude signal in the second frequency domain signal is taken as the fault frequency. The difference between the fault frequency and the first, second, or third order characteristic frequencies of the component is calculated, i.e., whether the fault frequency matches the first, second, or third order characteristic frequency of a rotating component. For example, for a bearing, the first order characteristic frequency is f1 = Z × (N / 60), where Z is the number of bearing balls and N is the rotational speed (rpm). The second and third order characteristic frequencies are typically 2 and 3 times the first order characteristic frequency, respectively. The second order characteristic frequency is f2 = 2 × f1, and the third order characteristic frequency is f3 = 3 × f1. If the difference between the fault frequency and any of f1, f2, or f3 is less than a preset difference (such as 3 Hz), the component is determined to be a faulty component.

[0048] Further, in one embodiment, before calculating the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the first-order characteristic frequency, second-order characteristic frequency, and third-order characteristic frequency of the component for each component, the method includes: For each component, the first-order characteristic frequency of the component is calculated based on the component's fundamental frequency and rotational speed, where the rotational speed is determined based on the vehicle speed at the time of signal acquisition. The second-order and third-order characteristic frequencies of the components are calculated based on their first-order characteristic frequencies.

[0049] In this embodiment, the rotational speed is first calculated based on the vehicle speed (obtained via the vehicle's CAN bus). Then, the first-order characteristic frequency is calculated based on the fundamental frequencies of the components (such as the number of balls in a bearing, the number of teeth in a gear, etc.). The formula for calculating the first-order characteristic frequency is: First-order characteristic frequency = Fundamental frequency × (Rotational speed / 60). The second-order and third-order characteristic frequencies are typically 2 and 3 times the first-order characteristic frequency, respectively: Second-order characteristic frequency = 2 × First-order characteristic frequency, and third-order characteristic frequency = 3 × First-order characteristic frequency. The characteristic frequencies of components are directly related to their fundamental frequencies and rotational speeds. By reasonably calculating the characteristic frequencies, accurate references can be provided for fault matching.

[0050] Furthermore, in one embodiment, calculating the first-order characteristic frequency of the component based on its fundamental frequency and rotational speed includes: The first-order characteristic frequency of the component is calculated using Formula 2 based on the component's fundamental frequency and rotational speed. Formula 2 is as follows: The first-order characteristic frequency of a component = the fundamental frequency of the component × (rotation speed / 60).

[0051] In this embodiment, taking a bearing as an example, the fundamental frequency is the number of bearing balls Z, and the rotational speed N is the engine speed (rpm) corresponding to the current vehicle speed. Then, the first-order characteristic frequency f1 = Z × (N / 60). This formula takes into account the linear relationship between rotational speed and characteristic frequency. The bearing fault characteristic frequency is proportional to the rotational speed; the higher the rotational speed, the higher the characteristic frequency.

[0052] Furthermore, in one embodiment, the period during which the powertrain makes abnormal noise is determined based on the sound signals collected by the microphone.

[0053] In this embodiment, abnormal noises from the vehicle powertrain are usually accompanied by abnormal sound signals. By analyzing these sound signals, the timing of the abnormal noise can be determined earlier and more accurately, avoiding the lag in vibration signal acquisition. Specifically, microphones can be placed near the powertrain to collect sound signals. By analyzing the time and frequency domain characteristics of the sound signals, the timing of the abnormal noise can be determined, and the corresponding vibration signal can be identified. For example, when the amplitude of the sound signal exceeds a preset threshold or an abnormal peak appears in the spectrum, the corresponding time period is determined to be the time period of the abnormal noise.

[0054] Furthermore, in one embodiment, the preset amplitude is a preset multiple of the average amplitude of the vibration signal during the period when the powertrain is free of abnormal noise.

[0055] In this embodiment, the amplitude of normal vibration signals is usually relatively stable, while the amplitude of fault vibration signals will increase significantly. By setting a reasonable multiple, normal vibration and fault vibration can be effectively distinguished, avoiding misjudging normal vibration signals as fault signals and improving the accuracy of fault frequency range determination. The preset amplitude is, for example, 1.5 times the average amplitude of vibration signals during periods when the powertrain has no abnormal noise. The preset multiple can be derived based on experimental data.

[0056] Secondly, embodiments of this application also provide a device for determining faulty components.

[0057] In one embodiment, reference is made to Figure 8 , Figure 8 This is a functional module diagram of an embodiment of the faulty component determination device of this application, as shown below. Figure 8 As shown, the faulty component identification device includes: The acquisition module 10 is used to acquire vibration signals during the period when abnormal noise occurs in the powertrain; The first conversion module 20 is used to convert the vibration signal into a first frequency domain signal by performing a Fourier transform, and to determine the fault frequency range based on the amplitude of the first frequency domain signal. The decomposition module 30 is used to perform wavelet packet decomposition on the vibration signal to obtain multiple sub-band signals, wherein each sub-band signal is in a different frequency band, and the sub-band signals in the fault frequency band are combined to form the fault vibration signal. The calculation module 40 is used to perform empirical mode decomposition on the fault vibration signal to obtain multiple intrinsic mode functions, and to calculate the correlation coefficient between each intrinsic mode function and the fault vibration signal. The second conversion module 50 is used to perform Hilbert transform on the intrinsic mode functions with correlation coefficients greater than preset coefficients, solve for the envelope signal of the intrinsic mode functions, and perform Fourier transform on the envelope signal to obtain the second frequency domain signal. The comparison module 60 is used to calculate the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the characteristic frequency of each component of the powertrain, and to identify the component whose difference is less than a preset difference as the faulty component.

[0058] Furthermore, in one embodiment, the determination of the fault frequency band based on the amplitude of the first frequency domain signal is used for: The first frequency domain signal is divided into multiple frequency intervals; For each frequency range, if the average signal amplitude within the frequency range is greater than the preset amplitude, then the frequency range is designated as the fault frequency segment.

[0059] Further, in one embodiment, the correlation coefficient is the Pearson correlation coefficient, and the calculation of the correlation coefficient between each intrinsic mode function and the fault vibration signal is used for: The Pearson correlation coefficient between each intrinsic mode function and the fault vibration signal is calculated using Formula 1, which is: ; Where C is the Pearson correlation coefficient, and n is the total number of signal sampling points. This represents the value of each intrinsic mode function at the i-th sampling point. This represents the mean of all sampled points for each intrinsic mode function. Let be the value of the fault vibration signal at the i-th sampling point. This is the mean value of all sampling points of the fault vibration signal.

[0060] Furthermore, in one embodiment, the Hilbert transform of the intrinsic mode functions and the solution of the envelope signal of the intrinsic mode functions are used for: By performing a Hilbert transform on the intrinsic mode functions, the orthogonal components of the intrinsic mode functions are obtained; The eigenmode functions and orthogonal components are combined to obtain the analytic signal; The envelope signal of the intrinsic mode function is obtained by taking the modulus of the analytic signal.

[0061] Furthermore, in one embodiment, the characteristic frequencies of the component include first-order characteristic frequencies, second-order characteristic frequencies, and third-order characteristic frequencies. The comparison module 60 is used for: For each component, calculate the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the first-order characteristic frequency, second-order characteristic frequency and third-order characteristic frequency of the component; If any difference is less than the preset difference, the component will be considered a faulty component.

[0062] Furthermore, in one embodiment, the faulty component determination device further includes a characteristic frequency calculation module, used for: For each component, the first-order characteristic frequency of the component is calculated based on the component's fundamental frequency and rotational speed, where the rotational speed is determined based on the vehicle speed at the time of signal acquisition. The second-order and third-order characteristic frequencies of the components are calculated based on their first-order characteristic frequencies.

[0063] Furthermore, in one embodiment, the first-order characteristic frequency of the component calculated based on its fundamental frequency and rotational speed is used for: The first-order characteristic frequency of the component is calculated using Formula 2 based on the component's fundamental frequency and rotational speed. Formula 2 is as follows: The first-order characteristic frequency of a component = the fundamental frequency of the component × (rotation speed / 60).

[0064] Furthermore, in one embodiment, the period during which the powertrain makes abnormal noise is determined based on the sound signals collected by the microphone.

[0065] Furthermore, in one embodiment, the preset amplitude is a preset multiple of the average amplitude of the vibration signal during the period when the powertrain is free of abnormal noise.

[0066] The functions of each module in the above-mentioned faulty component determination device correspond to the steps in the above-mentioned faulty component determination method embodiment, and their functions and implementation processes will not be described in detail here.

[0067] Thirdly, embodiments of this application provide a device for determining faulty components.

[0068] Reference Figure 9 , Figure 9 This is a schematic diagram of the hardware structure of the faulty component determination device involved in the embodiments of this application. In the embodiments of this application, the faulty component determination device may include a processor, a memory, a communication interface, and a communication bus.

[0069] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0070] Communication interfaces include input / output (I / O) interfaces, physical interfaces, and logical interfaces used for interconnecting devices within the fault-finding device, as well as interfaces used for interconnecting the fault-finding device with other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0071] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0072] The processor can be a general-purpose processor, which can call a faulty component determination program stored in memory and execute the faulty component determination method provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the faulty component determination program is called can be referred to in the various embodiments of the faulty component determination method of this application, and will not be repeated here.

[0073] Those skilled in the art will understand that Figure 9 The hardware structure shown does not constitute a limitation of this application and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0074] Fourthly, embodiments of this application also provide a readable storage medium.

[0075] The present application has a fault component determination program stored on a readable storage medium, wherein when the fault component determination program is executed by a processor, it implements the steps of the fault component determination method as described above.

[0076] The method implemented when the faulty component determination procedure is executed can be referred to in various embodiments of the faulty component determination method of this application, and will not be repeated here.

[0077] It should be noted that the sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0078] The terms "comprising" and "having," and any variations thereof, in the specification, claims, and accompanying drawings of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus. The terms "first," "second," and "third," etc., are used to distinguish different objects, etc., and do not indicate a sequence, nor do they limit "first," "second," and "third" to different types.

[0079] In the description of the embodiments of this application, terms such as "exemplary," "for example," or "for instance" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as "exemplary," "for example," or "for instance" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of terms such as "exemplary," "for example," or "for instance" is intended to present the relevant concepts in a concrete manner.

[0080] In the description of the embodiments of this application, unless otherwise stated, " / " means "or". For example, A / B can mean A or B. The "and / or" in the text is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of this application, "multiple" means two or more.

[0081] In some processes described in the embodiments of this application, multiple operations or steps are included in a specific order. However, it should be understood that these operations or steps may not be executed in the order they appear in the embodiments of this application, or they may be executed in parallel. The sequence number of the operation is only used to distinguish different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed sequentially or in parallel, and these operations or steps may be combined.

[0082] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of this application.

[0083] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A method for determining faulty components, characterized in that, The method for determining the faulty component includes: Vibration signals were collected during the periods when abnormal noises occurred in the powertrain; The vibration signal is converted into a first frequency domain signal by Fourier transform, and the fault frequency range is determined based on the amplitude of the first frequency domain signal. Wavelet packet decomposition of the vibration signal yields multiple sub-band signals, each of which is located in a different frequency band. The sub-band signals located in the fault frequency band are combined to form the fault vibration signal. The fault vibration signal is subjected to empirical mode decomposition to obtain multiple intrinsic mode functions, and the correlation coefficient between each intrinsic mode function and the fault vibration signal is calculated. For intrinsic mode functions with correlation coefficients greater than preset coefficients, Hilbert transform is performed on the intrinsic mode functions, and the envelope signal of the intrinsic mode functions is solved. Fourier transform is then performed on the envelope signal to obtain the second frequency domain signal. Calculate the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the characteristic frequency of each component of the powertrain, and identify the component whose difference is less than a preset difference as the faulty component.

2. The method for determining faulty components as described in claim 1, characterized in that, The determination of the fault frequency band based on the amplitude of the first frequency domain signal includes: The first frequency domain signal is divided into multiple frequency intervals; For each frequency range, if the average signal amplitude within the frequency range is greater than the preset amplitude, then the frequency range is designated as the fault frequency segment.

3. The method for determining faulty components as described in claim 1, characterized in that, The correlation coefficient is the Pearson correlation coefficient, and the calculation of the correlation coefficient between each intrinsic mode function and the fault vibration signal includes: The Pearson correlation coefficient between each intrinsic mode function and the fault vibration signal is calculated using Formula 1, which is: ； Where C is the Pearson correlation coefficient, and n is the total number of signal sampling points. This represents the value of each intrinsic mode function at the i-th sampling point. This represents the mean of all sampled points for each intrinsic mode function. Let be the value of the fault vibration signal at the i-th sampling point. This is the mean value of all sampling points of the fault vibration signal.

4. The method for determining faulty components as described in claim 1, characterized in that, The process of performing a Hilbert transform on the intrinsic mode functions and solving for the envelope signal of the intrinsic mode functions includes: By performing a Hilbert transform on the intrinsic mode functions, the orthogonal components of the intrinsic mode functions are obtained; The eigenmode functions and orthogonal components are combined to obtain the analytic signal; The envelope signal of the intrinsic mode function is obtained by taking the modulus of the analytic signal.

5. The method for determining faulty components as described in claim 1, characterized in that, The characteristic frequencies of the components include first-order characteristic frequencies, second-order characteristic frequencies, and third-order characteristic frequencies. The step of calculating the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the characteristic frequency of each component in the powertrain, and identifying components with a difference less than a preset difference as faulty components, includes: For each component, calculate the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the first-order characteristic frequency, second-order characteristic frequency and third-order characteristic frequency of the component; If any difference is less than the preset difference, the component will be considered a faulty component.

6. The method for determining faulty components as described in claim 5, characterized in that, Before calculating the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the first-order characteristic frequency, second-order characteristic frequency, and third-order characteristic frequency of the component for each component, the following steps are included: For each component, the first-order characteristic frequency of the component is calculated based on the component's fundamental frequency and rotational speed, where the rotational speed is determined based on the vehicle speed at the time of signal acquisition. The second-order and third-order characteristic frequencies of the components are calculated based on their first-order characteristic frequencies.

7. The method for determining faulty components as described in claim 6, characterized in that, The first-order characteristic frequency of the component, calculated based on its fundamental frequency and rotational speed, includes: The first-order characteristic frequency of the component is calculated using Formula 2 based on the component's fundamental frequency and rotational speed. Formula 2 is as follows: The first-order characteristic frequency of a component = the fundamental frequency of the component × (rotation speed / 60).

8. The method for determining faulty components as described in claim 1, characterized in that, The timing of abnormal noises in the powertrain was determined based on sound signals collected by the microphone.

9. The method for determining faulty components as described in claim 2, characterized in that, The preset amplitude is a preset multiple of the average amplitude of the vibration signal during the period when the powertrain is free of abnormal noise.

10. A device for identifying faulty components, characterized in that, The faulty component identification device includes: The acquisition module is used to collect vibration signals during periods when abnormal noises occur in the powertrain; The first conversion module is used to convert the vibration signal into a first frequency domain signal by performing a Fourier transform, and to determine the fault frequency range based on the amplitude of the first frequency domain signal. The decomposition module is used to perform wavelet packet decomposition on the vibration signal to obtain multiple sub-band signals, where each sub-band signal is in a different frequency band, and the sub-band signals in the fault frequency band are combined to form the fault vibration signal. The calculation module is used to perform empirical mode decomposition on the fault vibration signal to obtain multiple intrinsic mode functions, and calculate the correlation coefficient between each intrinsic mode function and the fault vibration signal; The second conversion module is used to perform Hilbert transform on the intrinsic mode functions with correlation coefficients greater than preset coefficients, solve for the envelope signal of the intrinsic mode functions, and perform Fourier transform on the envelope signal to obtain the second frequency domain signal. The comparison module is used to calculate the difference between the frequency corresponding to the largest amplitude signal in the second frequency domain signal and the characteristic frequency of each component of the powertrain, and to identify the component whose difference is less than a preset difference as the faulty component.

11. The faulty component identification device as described in claim 10, characterized in that, The method of determining the fault frequency band based on the amplitude of the first frequency domain signal is used for: The first frequency domain signal is divided into multiple frequency intervals; For each frequency range, if the average signal amplitude within the frequency range is greater than the preset amplitude, then the frequency range is designated as the fault frequency segment.

12. The faulty component identification device as described in claim 10, characterized in that, The correlation coefficient is the Pearson correlation coefficient, and the calculation of the correlation coefficient between each intrinsic mode function and the fault vibration signal is used for: The Pearson correlation coefficient between each intrinsic mode function and the fault vibration signal is calculated using Formula 1, which is: ； Where C is the Pearson correlation coefficient, and n is the total number of signal sampling points. This represents the value of each intrinsic mode function at the i-th sampling point. This represents the mean of all sampled points for each intrinsic mode function. Let be the value of the fault vibration signal at the i-th sampling point. This is the mean value of all sampling points of the fault vibration signal.

13. The faulty component identification device as described in claim 10, characterized in that, The Hilbert transform of the intrinsic mode functions and the solution of the envelope signal of the intrinsic mode functions are used for: By performing a Hilbert transform on the intrinsic mode functions, the orthogonal components of the intrinsic mode functions are obtained; The eigenmode functions and orthogonal components are combined to obtain the analytic signal; The envelope signal of the intrinsic mode function is obtained by taking the modulus of the analytic signal.

14. A faulty component identification device, characterized in that, The faulty component determination device includes a processor, a memory, and a faulty component determination program stored in the memory and executable by the processor, wherein when the faulty component determination program is executed by the processor, it implements the steps of the faulty component determination method as described in any one of claims 1 to 9.

15. A readable storage medium, characterized in that, The readable storage medium stores a faulty component determination program, wherein when the faulty component determination program is executed by a processor, it implements the steps of the faulty component determination method as described in any one of claims 1 to 9.

Images