Rolling bearing fault diagnosis method based on piezoelectric stress wave signal active sensing

Abstract

The invention discloses a rolling bearing fault diagnosis method based on piezoelectric stress wave signal active sensing, and relates to the technical field of rolling bearing fault diagnos.The rolling bearing fault diagnosis method comprises the steps that under the working state of a rolling bearing, a pair of sheet type piezoelectric ceramic sensors are arranged at the monitoring position of the rolling bearing; one sheet-type piezoelectric ceramic sensor is used for exciting a high-frequency stress wave signal, the other sheet-type piezoelectric ceramic sensor is used for receiving a high-frequency stress wave modulation signal obtained after the high-frequency stress wave signal is propagated through a rolling bearing, spectral analysis is carried out on the received high-frequency stress wave modulation signal based on the technologies such as FFT and envelope spectrum, and the high-frequency stress wave modulation signal is obtained. Fault diagnosis and state monitoring of the rolling bearing are completed by identifying fault frequency spectrum features, the method is more sensitive to early damage, the health state of the rolling bearing can be identified earlier, and the method has important significance on condition-based maintenance of various devices.

Description

Technical Field

[0001] This invention relates to the field of rolling bearing fault diagnosis technology, specifically to a rolling bearing fault diagnosis method and system based on active sensing of piezoelectric stress wave signals. Background Technology

[0002] For various rotating machinery such as aircraft engines, turbines, high-speed trains, and wind power, health monitoring of rolling bearings, one of their key components, can effectively prevent abnormal operating conditions caused by various faults. In the past, inspectors mostly used visual inspection methods to periodically assess the health status and fault conditions of rolling bearings. However, this method requires disassembly and equipment shutdown, as well as a large amount of manpower and resources, significantly increasing operation and maintenance costs. Furthermore, manual inspection relies heavily on personnel experience, resulting in poor reliability.

[0003] Furthermore, related fields have proposed methods for monitoring and diagnosing the condition of rolling bearings based on vibration and stress wave signals to avoid the problems caused by manual periodic inspections. However, it is worth noting that vibration signals are not sensitive to early damage to rolling bearings and are significantly affected by environmental noise; stress wave signals attenuate rapidly and are easily submerged by noise, significantly reducing the monitoring and diagnostic effectiveness. At the same time, most of the above methods rely on machine learning and deep learning technologies, requiring a large number of fault signals for diagnostic model training, and real-time performance is difficult to guarantee (because the deployment of artificial intelligence algorithms significantly consumes the memory resources of the monitoring and diagnostic system), thus presenting significant limitations in practical engineering applications. On the other hand, methods for monitoring and diagnosing the condition of rolling bearings often require placing sensors at each bearing monitoring location. However, due to size constraints and practical operating conditions, vibration sensors and stress wave sensors often need to be placed at a distance from the bearing being monitored, further reducing the monitoring and diagnostic effectiveness.

[0004] Therefore, proposing a new fault diagnosis method and system to realize the condition monitoring and fault diagnosis of rolling bearings is an urgent problem to be solved by those skilled in the art. Summary of the Invention

[0005] To address the shortcomings of existing technologies, a method and system for diagnosing rolling bearing faults based on active sensing of piezoelectric stress wave signals has been invented, enabling condition monitoring and fault diagnosis of rolling bearings.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a rolling bearing fault diagnosis method based on active sensing of piezoelectric stress wave signals, comprising the following steps: Step 1: With the rolling bearing in operation, place a pair of plate-type piezoelectric ceramic sensors at the monitoring position of the rolling bearing. One of the plate-type piezoelectric ceramic sensors is used to excite a high-frequency stress wave signal with a frequency of [frequency value missing]. Another chip-type piezoelectric ceramic sensor is used to receive the high-frequency stress wave modulated signal after the high-frequency stress wave signal propagates through the rolling bearing. ; Step 2: Identify the main types of rolling bearing failures, including at least inner ring failures, outer ring failures, and rolling element failures; Step 3: Modulate the received high-frequency stress wave signal Based on techniques such as FFT and envelope spectrum, spectrum analysis is carried out to identify fault spectrum characteristics and complete the fault diagnosis and condition monitoring of rolling bearings. Preferably, the characteristic frequencies associated with inner ring faults are: ; The characteristic frequencies associated with outer ring faults are: ; Regarding the characteristic frequencies associated with rolling element failures: ; Where G is the number of rolling elements, d is the diameter of the rolling elements, and D is the average diameter of the rolling bearing. The radial contact angle, The rotational speed of the shaft.

[0007] Preferably, in step 3, the high-frequency stress wave modulation signal S m The specific methods for conducting spectrum analysis are as follows: The passband range of the bandpass filter is set to 90kHz-110kHz. According to the Nyquist sampling theorem, f is set... S =256kHz, select N=8192, confirm sampling time T=N / f S =8192 / 256000=0.032s; Different high-frequency stress wave modulation signals S associated with different time variables m Calibrated as S m (k), the discrete FFT transform of the time-domain signal Sm(k) is given by the following formula: , where k = 0, 1, 2, ..., N-1, calculate the frequency domain assigned spectrum |Sm(k)|, where the frequency axis f k =k×f S / N; Generate a discrete spectrum plot of frequency axis f (0-128 kHz) and amplitude |Sm(f)|, with the horizontal axis representing frequency and the vertical axis representing amplitude; The specific methods for fault diagnosis and condition monitoring of rolling bearings are as follows: Based on the generated discrete spectrum, the frequency is selected as... The numerical range is ±10kHz, and a frequency is randomly selected from within this range as the verification frequency. The amplitude associated with the verification frequency is calibrated as F.q Where q represents the different check frequencies within the corresponding numerical range, and the confirmed check frequency is denoted as P. q ; Then, based on the determined inner ring fault characteristic frequency Outer ring fault characteristic frequency and rolling element failure characteristic frequency Lock the search range segment (P) q ± ), (P q ± ) and (P q ± ); From the confirmed search range segment (P) q ± In ), the amplitude F is identified. q Are there any abnormal amplitude points on both sides? The amplitude range associated with these abnormal amplitude points is (F). q If there are abnormal amplitude points on both sides, an inner ring fault signal is directly generated and displayed. If there is no abnormal amplitude point, other frequencies are selected from the numerical range as the verification frequency and the inner ring fault is verified to identify whether an inner ring fault signal is generated. If it is not generated, other frequencies are continuously selected as the verification frequency, and so on, until all frequency points in the numerical range are selected as verification frequencies. Using the same confirmation method as that used to generate inner ring fault signals, the search range segment (P) is identified. q ± In ), the amplitude F is identified. q If there are abnormal amplitude points on both sides, an outer ring fault signal will be generated and displayed directly. If not, other frequencies will be selected as verification frequencies using the same selection method and continuous verification will be performed. For the confirmed search range segment (P) q ± The system identifies abnormal amplitude points. If an abnormal amplitude point exists, a rolling element fault signal is generated and displayed directly. If no abnormal amplitude point exists, the system continuously selects a verification frequency for abnormal verification.

[0008] Preferably, the rolling bearing fault diagnosis system based on active sensing of piezoelectric stress wave signals includes: A pair of plate-type piezoelectric ceramic sensors are arranged at the monitoring position of the rolling bearing. The plate-type piezoelectric ceramic sensors are respectively connected to the excitation source and the data acquisition and analysis system via cables. A signal power amplifier is also connected between the plate-type piezoelectric ceramic sensor that emits high-frequency stress wave signal and the excitation source. Preferably, in the above steps, a signal power amplifier is used to amplify the electrical signal emitted by the excitation source, and then the plate-type piezoelectric ceramic sensor converts the amplified electrical signal into a high-frequency stress wave signal based on the inverse piezoelectric effect of the piezoelectric material. Preferably, in the above steps, after the high-frequency stress wave signal propagates through the rolling bearing, another plate-type piezoelectric ceramic sensor, based on the inverse piezoelectric effect of the piezoelectric material, converts the received high-frequency stress wave modulation signal into an electrical signal, which is then collected and stored by the data acquisition and analysis system and the host computer. Preferably, the received high-frequency stress wave modulation signal is filtered and noise-reduced.

[0009] Preferably, spectral analysis is performed on the high-frequency stress wave modulation signal obtained after filtering and noise reduction, and the fault diagnosis and identification of the rolling bearing are completed based on the fault spectral characteristics.

[0010] Preferably, the stimulus source and data acquisition and analysis system can be replaced by a system that has both stimulus and data acquisition and analysis capabilities.

[0011] This invention provides a method for fault diagnosis of rolling bearings based on active sensing of piezoelectric stress wave signals. Compared with existing technologies, it has the following advantages: (1) Compared with passive methods such as vibration and stress wave signals, the rolling bearing fault diagnosis method based on active sensing of piezoelectric stress wave signals is more sensitive to early damage and can identify the health status of rolling bearings earlier, which is of great significance for condition-based maintenance of various equipment. (2) Compared with existing technologies, the rolling bearing fault diagnosis method based on active sensing of piezoelectric stress wave signals does not require feature extraction and analysis of signals based on machine learning, deep learning and other models, thus making up for the problem caused by insufficient fault signals in engineering practice. (3) Compared with vibration and stress wave sensors, the chip-type piezoelectric ceramic sensor used in this invention is small in size and easy to install and arrange, effectively solving the problem of online monitoring and fault diagnosis of rolling bearings in a limited space. Attached Figure Description

[0012] Figure 1 A flowchart of a rolling bearing fault diagnosis method based on active sensing of piezoelectric stress wave signals provided by the present invention; Figure 2 This is a schematic diagram of a rolling bearing fault diagnosis system based on active sensing of piezoelectric stress wave signals, provided by the present invention. Detailed Implementation

[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] First Embodiment Please see Figure 1 This application provides a method for diagnosing rolling bearing faults based on active sensing of piezoelectric stress wave signals, including the following steps: Step 1: Select bearing model 61805 rolling bearing. Arrange a pair of plate-type piezoelectric ceramic sensors at the monitoring position of the rolling bearing. One plate-type piezoelectric ceramic sensor is used to excite a high-frequency stress wave signal with a frequency of 100kHz; the other plate-type piezoelectric ceramic sensor is used to receive the high-frequency stress wave modulation signal S after the high-frequency stress wave signal propagates through the rolling bearing. m Mechanical waves generated by a sheet-type piezoelectric ceramic sensor based on the inverse piezoelectric effect propagate in the rolling bearing material in the form of elastic stress. The frequency is typically tens to hundreds of kilohertz; here, 100 kHz is chosen. Unlike passively received environmental vibrations or stress waves, actively excited stress waves can directionally penetrate bearing components and are more sensitive to early microscopic damage (such as surface cracks and wear). After propagating through the rolling bearing, the high-frequency stress wave generates a modulated received signal due to the action of the faulty component. The periodic impact or contact deformation of the faulty component (such as inner ring damage) modulates the amplitude, frequency, or phase of the high-frequency stress wave, forming a composite signal containing fault characteristics. The local stiffness change caused by the fault alters the stress wave propagation speed, leading to a frequency shift in the received signal, forming a signal with a frequency of f0. e Centered on (f) e ±nf 故障 The spectral characteristics of the sidebands (n=1, 2, 3, ...) are as follows: High-frequency stress wave signal (generated by the excitation source) → propagates through the bearing components → is modulated when encountering a fault → forming a high-frequency stress wave modulation signal S. m (Receive data from sensors); Step 2: Confirm the bearing parameters of the current rolling bearing, and based on the confirmed bearing parameters, confirm the fault characteristic frequencies associated with the current rolling bearing. These bearing parameters include the number of rolling elements G, the rolling element diameter d, the average diameter of the rolling bearing D, and the radial contact angle. , The rotational speed of the shaft is given, and the specific method for confirming the fault characteristic frequency is as follows: use: , as well as Confirm the characteristic frequencies of inner ring faults in sequence. Outer ring fault characteristic frequency and rolling element failure characteristic frequency ; The main failure types of the 61805 rolling bearing are identified, including at least inner ring failure, outer ring failure, and rolling element failure. Further analysis using bearing parameters and formulas allows for the calculation of the corresponding failure characteristic frequencies. The 61805 bearing parameters include: the number of rolling elements. ; Rolling element diameter Average diameter of rolling bearings Radial contact angle At rotational speed The fault characteristic frequencies at 2000 rpm can be calculated as follows: ; ; ; Step 3: Based on FFT and envelope spectrum, analyze the received high-frequency stress wave modulation signal S m Conduct spectrum analysis to identify fault spectrum characteristics and complete fault diagnosis and condition monitoring of rolling bearings; The specific methods for conducting spectrum analysis are as follows: The passband range of the bandpass filter is set to 90 kHz-110 kHz (based on the excitation frequency f). e Centered at 100 kHz (±10% frequency band), filtering noise below 90 kHz and above 110 kHz, a wavelet denoising algorithm is used to decompose the filtered signal into multiple layers (e.g., 5 layers of db4 wavelet decomposition) to remove white noise interference and retain stress wave modulation characteristics. According to the Nyquist sampling theorem, let f S =256 kHz, choose N=8192 (a power of 2 for easier FFT calculation), and confirm the sampling time T=N / f S =8192 / 256000=0.032s; Different high-frequency stress wave modulation signals S associated with different time variables m Calibrated as S m (k), the discrete FFT transform of the time-domain signal Sm(k) is given by the following formula: , where k = 0, 1, 2, ..., N-1, calculate the frequency domain assigned spectrum |Sm(k)|, where the frequency axis f k =k×f S / N, for example: the first point k=0: DC component f=0 Hz, the k=3200th point: f=3200×256000 / 8192=100 kHz (corresponding to excitation frequency 100kHz); Generate a discrete spectrum of frequency axis f (0-128 kHz) and amplitude |Sm(f)|, with the horizontal axis representing frequency and the vertical axis representing amplitude (dB or linear value). The specific methods for fault diagnosis and condition monitoring of rolling bearings are as follows: Based on the generated discrete spectrum, the frequency is selected as... The numerical range is ±10kHz, and a frequency is randomly selected from within this range as the verification frequency. The amplitude associated with the verification frequency is calibrated as F. q Where q represents the different check frequencies within the corresponding numerical range, and the confirmed check frequency is denoted as P. q ; Then, based on the determined inner ring fault characteristic frequency Outer ring fault characteristic frequency and rolling element failure characteristic frequency Lock the search range segment (P) q ± ), (P q ± ) and (P q ± ); From the confirmed search range segment (P) q ± In ), the amplitude F is identified. q Are there any abnormal amplitude points on both sides? The amplitude range associated with these abnormal amplitude points is (F). q If there are abnormal amplitude points on both sides, an inner ring fault signal is directly generated and displayed. If there is no abnormal amplitude point, other frequencies are selected from the numerical range as the verification frequency and the inner ring fault is verified to identify whether an inner ring fault signal is generated. If it is not generated, other frequencies are continuously selected as the verification frequency, and so on, until all frequency points in the numerical range are selected as verification frequencies. Then, using the same confirmation method as generating the inner ring fault signal, start from the confirmed search range segment (P) q ± In ), the amplitude F is identified. q If there are abnormal amplitude points on both sides, an outer ring fault signal will be generated and displayed directly. If not, other frequencies will be selected as verification frequencies using the same selection method and continuous verification will be performed. Then, for the confirmed search range segment (P) q ± The system identifies abnormal amplitude points. If an abnormal amplitude point exists, a rolling element fault signal is generated and displayed directly. If no abnormal amplitude point exists, the system continuously selects a verification frequency for abnormal verification.

[0015] If a fault spectrum characteristic frequency exists This indicates an inner ring fault; if a fault spectrum characteristic frequency exists... This indicates an outer ring fault; if a fault spectrum characteristic frequency exists... If so, it indicates a rolling element failure.

[0016] Second Embodiment Combination Figure 2 As shown, the rolling bearing fault diagnosis system based on active sensing of piezoelectric stress wave signals includes: 1-excitation source, 2-signal power amplifier, 3-chip piezoelectric ceramic sensor (excitation), 4-rolling bearing, 5-chip piezoelectric ceramic sensor (receiver), 6-data acquisition and analysis system, and 7-host computer.

[0017] Positions 3 and 5 are placed at monitoring location 4. A 100 kHz sinusoidal signal is excited by position 1, amplified by position 2, and then used to excite position 3 to emit a 100 kHz high-frequency stress wave signal. The high-frequency stress wave signal propagates through position 4 and is received by position 5. Positions 6 and 7 then perform acquisition, filtering, noise reduction, and storage. Spectral analysis is performed on the high-frequency stress wave modulation signal obtained after filtering and noise reduction. Based on the fault spectrum characteristics, fault diagnosis and identification of the rolling bearing are completed.

[0018] Preferably, the stimulus source and data acquisition and analysis system can be replaced by a system that has both stimulus and data acquisition and analysis capabilities.

[0019] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0020] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for fault diagnosis of rolling bearings based on active sensing of piezoelectric stress wave signals, characterized in that, Includes the following steps: Step 1: With the rolling bearing in operation, place a pair of plate-type piezoelectric ceramic sensors at the monitoring position of the rolling bearing. One of the plate-type piezoelectric ceramic sensors is used to excite a high-frequency stress wave signal with a frequency of [frequency value missing]. Another chip-type piezoelectric ceramic sensor is used to receive the high-frequency stress wave modulated signal after the high-frequency stress wave signal propagates through the rolling bearing. ; Step 2: Identify the main types of rolling bearing failures, including at least inner ring failures, outer ring failures, and rolling element failures; Step 3: Modulate the received high-frequency stress wave signal Based on techniques such as FFT and envelope spectrum, spectral analysis is carried out to identify fault spectral characteristics and complete the fault diagnosis and condition monitoring of rolling bearings.

2. The rolling bearing fault diagnosis method based on active sensing of piezoelectric stress wave signals according to claim 1, characterized in that, In step 1, the high-frequency stress wave signal is generated by the excitation source, and the transformation of the high-frequency stress wave signal into a high-frequency stress wave modulation signal is as follows: high-frequency stress wave signal (generated by the excitation source) → propagates through the bearing component → is modulated when a fault occurs → forms a high-frequency stress wave modulation signal S. m .

3. The rolling bearing fault diagnosis method based on active sensing of piezoelectric stress wave signals according to claim 1, characterized in that, In step 2: The characteristic frequencies associated with inner-circle faults are: ; The characteristic frequencies associated with outer ring faults are: ; Regarding the characteristic frequencies associated with rolling element failures: ; Where G is the number of rolling elements, d is the diameter of the rolling elements, and D is the average diameter of the rolling bearing. The radial contact angle, The rotational speed of the shaft is denoted as .

4. The rolling bearing fault diagnosis method based on active sensing of piezoelectric stress wave signals according to claim 1, characterized in that, In step 3, the high-frequency stress wave modulation signal S m The specific methods for conducting spectrum analysis are as follows: The passband range of the bandpass filter is set to 90kHz-110kHz. According to the Nyquist sampling theorem, f is set... S =256kHz, select N=8192, confirm sampling time T=N / f S =8192 / 256000=0.032s; Different high-frequency stress wave modulation signals S associated with different time variables m Calibrated as S m (k), the discrete FFT transform of the time-domain signal Sm(k) is given by the following formula: , where k = 0, 1, 2, ..., N-1, calculate the frequency domain assigned spectrum |Sm(k)|, where the frequency axis f k =k×f S / N; Generate a discrete spectrum of frequency axis f (0-128 kHz) and amplitude |Sm(f)|, with the horizontal axis representing frequency and the vertical axis representing amplitude.

5. The rolling bearing fault diagnosis method based on active sensing of piezoelectric stress wave signals according to claim 1, characterized in that, In step 3, the specific method for fault diagnosis and condition monitoring of the rolling bearing is as follows: Based on the generated discrete spectrum, the frequency is selected as... The numerical range is ±10kHz, and a frequency is randomly selected from within this range as the verification frequency. The amplitude associated with the verification frequency is calibrated as F. q Where q represents the different check frequencies within the corresponding numerical range, and the confirmed check frequency is denoted as P. q ; Then, based on the determined inner ring fault characteristic frequency Outer ring fault characteristic frequency and rolling element failure characteristic frequency Lock the search range segment (P) q ± ), (P q ± ) and (P q ± ); From the confirmed search range segment (P) q ± In ), the amplitude F is identified. q Are there any abnormal amplitude points on both sides? The amplitude range associated with these abnormal amplitude points is (F). q If there are abnormal amplitude points on both sides (±15dB), an inner ring fault signal is directly generated and displayed. If not, another frequency is selected from the numerical range as the verification frequency, and the inner ring fault is verified to identify whether an inner ring fault signal is generated. If not, another frequency is continuously selected as the verification frequency, and so on, until all frequency points in the numerical range are selected as verification frequencies.

6. The rolling bearing fault diagnosis method based on active sensing of piezoelectric stress wave signals according to claim 5, characterized in that, In step 3, the same confirmation method used to generate the inner ring fault signal is employed to check the confirmed search range segment (P). q ± In ), the amplitude F is identified. q If there are abnormal amplitude points on both sides, an outer ring fault signal is generated and displayed directly. If not, other frequencies are selected as verification frequencies using the same selection method and continuous verification is performed.

7. The rolling bearing fault diagnosis method based on active sensing of piezoelectric stress wave signals according to claim 5, characterized in that, In step 3, the confirmed search range segment (P) q ± Abnormal amplitude points are identified. If they exist, a rolling element fault signal is generated and displayed directly. If they do not exist, the verification frequency is continuously selected for abnormal verification.

8. A rolling bearing fault diagnosis system based on active sensing of piezoelectric stress wave signals, wherein the system operates according to any one of claims 1-7, characterized in that, include: Excitation source, signal power amplifier, chip piezoelectric ceramic excitation sensor, rolling bearing, chip piezoelectric ceramic receiving sensor, data acquisition and analysis system, host computer; Among them, a pair of sheet piezoelectric ceramic sensors are arranged at the rolling bearing monitoring position, namely a sheet piezoelectric ceramic excitation sensor and a sheet piezoelectric ceramic receiving sensor; The chip piezoelectric ceramic excitation sensor and the chip piezoelectric ceramic receiving sensor are connected to the excitation source and the data acquisition and analysis system via cables, respectively. A signal power amplifier is also connected between the chip piezoelectric ceramic excitation sensor that emits high-frequency stress wave signals and the excitation source.

9. The rolling bearing fault diagnosis system based on active sensing of piezoelectric stress wave signals according to claim 8, characterized in that, The signal power amplifier amplifies the electrical signal emitted by the excitation source, and then the plate-type piezoelectric ceramic sensor converts the amplified electrical signal into a high-frequency stress wave signal based on the inverse piezoelectric effect of the piezoelectric material. After the high-frequency stress wave signal propagates through the rolling bearing, another plate-type piezoelectric ceramic sensor, based on the inverse piezoelectric effect of the piezoelectric material, converts the received high-frequency stress wave modulation signal into an electrical signal, which is then collected and stored by the data acquisition and analysis system and the host computer. The host computer performs filtering and noise reduction processing on the received high-frequency stress wave modulation signal.

Images