Acoustic detection method and system for bearing fault based on fractional harmonic phase-locked amplification

Abstract

The application discloses a bearing fault acoustic detection method and system based on fractional harmonic phase-locked amplification, acoustics signals of a target motor are acquired, the shaft rotation frequency of the motor shaft is measured in real time, and is synchronized to a fractional phase-locked loop; according to the geometric parameters of the bearing, the bearing fault characteristic frequency is calculated, the bearing fault characteristic frequency is a non-integer multiple of the shaft rotation frequency, a reference signal synchronized to the bearing fault characteristic frequency is generated through the fractional phase-locked loop, and the acquired acoustics signals and the generated reference signal are detected in a phase-sensitive manner through a phase-locked amplifier; specifically, the acoustics signals and the reference signal are quadrature demodulated to generate in-phase components and quadrature components, direct current components are extracted through low-pass filtering, and then the vibration amplitude and phase information corresponding to the bearing fault characteristic frequency are calculated; according to the comparison between the vibration amplitude and a preset baseline, harmonic consistency and phase stability, the bearing fault type and severity are determined.

Description

Technical Field

[0001] This invention belongs to the field of mechanical fault diagnosis and signal processing technology, specifically relating to an acoustic detection method and system for bearing faults based on fractional harmonic lock-in amplification. Background Technology

[0002] In the industrial sector, electric motors are critical power equipment, and their operational stability and reliability are paramount. Bearings, as one of the core components of an electric motor, play a crucial role in supporting and guiding the rotor's rotation. According to relevant research, 30%-40% of motor failures are attributed to rolling bearing failures. Bearing failures typically manifest as surface damage to key components (such as the outer ring, inner ring, cage, or rolling elements). These failures not only cause abnormal vibration and noise during motor operation but also significantly reduce motor efficiency and lifespan, and in severe cases, can even lead to motor shutdown, resulting in substantial economic losses for industrial production.

[0003] Currently, there are numerous technologies for bearing fault detection. Vibration signal analysis (AVS), a relatively mature method, identifies faults by monitoring the amplitude and frequency of bearing vibrations. However, in practical applications, the installation location and orientation of the vibration sensor significantly affect the detection results, and it is easily affected by external vibration interference. Motor current characteristic analysis (MCSA) indirectly detects bearing faults by analyzing changes in motor current. However, factors such as motor load changes and power supply fluctuations can interfere with the current signal, affecting the accuracy of the detection results. Other technologies include thermal imaging, fiber optic sensing, magnetic field detection, and acoustic emission detection. While acoustic emission detection has high sensitivity for early-stage faults, it requires the installation of sensors inside the equipment, making it an invasive detection method, and it has high requirements for the detection environment.

[0004] In contrast, sound signal-based detection methods have unique advantages. Sound signals contain rich information about the equipment's operating status and propagate through the air, eliminating the need for direct contact with the equipment and enabling non-invasive detection. However, sound signal detection faces significant challenges in real-world industrial environments. Due to the presence of substantial background noise and vibrations from other equipment, target sound signals are often severely contaminated, resulting in extremely low signal-to-noise ratios (SNR). This makes it difficult for traditional frequency analysis methods based on Fourier transforms to accurately extract fault characteristic frequencies. For example, in factory workshops, the simultaneous operation of various mechanical equipment generates noise that superimposes, causing the weak sound signals generated by bearing failures to be submerged and difficult to detect and analyze effectively. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, achieve the goal of overcoming the signal-to-noise ratio limitation in sound signal detection, and improve the accuracy of bearing fault detection, this invention adopts the following technical solution:

[0006] The acoustic detection method for bearing faults based on fractional harmonic lock-in amplification includes the following steps:

[0007] Acquire acoustic signals from the mating shaft and bearings;

[0008] Obtain the shaft rotation frequency and synchronize it to the fractional phase-locked loop;

[0009] The bearing fault characteristic frequency is calculated based on the bearing's geometric parameters, and the bearing fault characteristic frequency is a non-integer multiple of the shaft's rotational frequency;

[0010] A reference signal synchronized with the bearing fault characteristic frequency is generated through the fractional phase-locked loop;

[0011] Phase-sensitive detection is performed on the acoustic signal and the reference signal using a lock-in amplifier; the acoustic signal and the reference signal are orthogonally demodulated to generate in-phase and quadrature components, and the DC component is extracted by low-pass filtering. After orthogonal demodulation, the vibration amplitude and phase information corresponding to the bearing fault characteristic frequency are obtained.

[0012] Based on the comparison of the vibration amplitude with the preset baseline, harmonic consistency, and phase stability, the bearing fault type and severity are determined.

[0013] Furthermore, the fractional phase-locked loop acquires the shaft rotation frequency, compares the shaft rotation frequency with the feedback signal using a phase frequency detector, and outputs an error signal proportional to the phase difference to drive a charge pump to generate a voltage signal. The voltage signal is filtered by a low-pass filter to remove high-frequency noise, generating a smooth DC control voltage. This voltage is used to adjust the output frequency of the voltage-controlled oscillator (VCO) to ensure that the feedback signal is phase-locked with the input signal. The VCO generates a high-frequency signal based on the DC control voltage, and its output frequency is linearly related to the control voltage. A fractional frequency divider dynamically divides the high-frequency signal to obtain the feedback signal, forming a closed-loop control to ensure that the output frequency is accurately locked to the target fractional harmonic. Finally, the fractional phase-locked loop outputs a reference signal that is phase-locked with the shaft rotation frequency.

[0014] Furthermore, the dynamic frequency division has a non-integer division ratio, and the frequency divider achieves fractional frequency division by alternately switching the average effect of integer frequency division values.

[0015] Furthermore, the bearing failure characteristic frequency is based on the acquired shaft rotation frequency and the number of bearing balls. Ball diameter Pitch circle diameter and contact angle It can be obtained through the following formula:

[0016]

[0017]

[0018]

[0019] in, This indicates the characteristic frequency of bearing inner ring failure. This indicates the characteristic frequency of bearing outer ring failure. Non-integer multiples of the characteristic frequency of bearing failure are represented by coefficients. Indicates the rotational frequency of the shaft. Represents angular frequency. The contact angle is the angle between the bearing and the outer raceway. The size of the contact angle changes the contact state and force distribution between the rolling elements and the raceway, which has a significant impact on the bearing's performance and failure frequency.

[0020] Furthermore, the lock-in amplifier multiplies the acoustic signal with the reference signal and its quadrature signal respectively to generate in-phase and quadrature components, and extracts the DC component through low-pass filtering to obtain the in-phase component. and orthogonal components ;

[0021] The composite amplitude is calculated using the same-direction and quadrature components. and phase difference :

[0022]

[0023]

[0024] Based on the synthesized amplitude and preset reference amplitude and reference phase The vibration amplitude and phase difference are calculated using the following formula:

[0025]

[0026]

[0027] in, Indicates the vibration amplitude. This indicates the phase difference.

[0028] Furthermore, the fractional phase-locked loop integrated frequency synthesizer is used to generate multi-order harmonic reference signals as an extension of the reference signal, which are used to detect different fault characteristic frequencies that may exist in the bearing. The calculation formula is as follows:

[0029]

[0030] in, This represents the multi-order harmonic reference signal. A preset positive integer representing the order of the harmonic;

[0031] The acoustic signal and the multi-order harmonic reference signal are orthogonally demodulated to generate multi-order in-phase components and multi-order quadrature components. Then, the multi-order DC components are extracted by low-pass filtering, and the multi-order harmonic vibration amplitude and multi-order phase information corresponding to the bearing fault characteristic frequency are calculated.

[0032] Furthermore, the bearing failure is determined as follows:

[0033] Under fault-free conditions, measure and record the baseline threshold of vibration amplitude at the characteristic frequencies of each bearing failure; acquire the vibration amplitude of the target bearing in real time. Phase difference And the amplitude of multi-order harmonic vibrations;

[0034] When the vibration amplitude is measured in real time A fault warning is triggered when the baseline threshold is exceeded.

[0035] Verify whether the amplitude of multi-order harmonic vibrations corresponding to the characteristic frequency of bearing failure increases synchronously;

[0036] Calculate phase difference Standard deviation ,like If the signal is less than 0.1 rad, it is confirmed that the signal originates from the target bearing.

[0037] The fault location is determined based on the characteristic frequency of the abnormal vibration amplitude, and the severity is classified according to the multiple by which the vibration amplitude exceeds the baseline; wherein, the multiple is positively correlated with the severity.

[0038] A bearing fault acoustic detection system based on fractional harmonic phase-locked loop amplification includes a fractional phase-locked loop, a phase-locked amplifier, and a fault diagnosis module.

[0039] The fractional phase-locked loop generates a reference signal synchronized with the bearing fault characteristic frequency based on the acquired shaft rotation frequency; wherein the bearing fault characteristic frequency is calculated based on the bearing's geometric parameters and is a non-integer multiple of the shaft rotation frequency;

[0040] The lock-in amplifier performs phase-sensitive detection on the acoustic signals of the shaft and bearing and the corresponding reference signals; it performs quadrature demodulation on the acoustic signals and the reference signals to generate in-phase and quadrature components, then extracts the DC component through low-pass filtering, and after demodulation by the lock-in amplifier, obtains the vibration amplitude and phase information corresponding to the bearing fault characteristic frequency;

[0041] The fault diagnosis module determines the type and severity of bearing faults based on the comparison of the vibration amplitude with a preset baseline, harmonic consistency, and phase stability.

[0042] Furthermore, the fractional phase-locked loop includes a phase-frequency detector, a charge pump, a low-pass filter, and a fractional frequency divider.

[0043] The phase frequency detector compares the acquired shaft rotation frequency with the feedback signal in phase and outputs an error signal proportional to the phase difference to drive the charge pump to generate a voltage signal.

[0044] The low-pass filter filters the voltage signal, removes high-frequency noise, and generates a smooth DC control voltage. This voltage is used to adjust the output frequency of the voltage-controlled oscillator (VCO) to ensure that the feedback signal is phase-locked with the input signal.

[0045] The voltage-controlled oscillator generates a high-frequency signal based on the DC control voltage, and its output frequency is linearly related to the control voltage.

[0046] The fractional frequency divider dynamically divides the high-frequency signal to obtain the feedback signal, forming a closed-loop control to ensure that the output frequency is accurately locked to the target fractional harmonic, so that the fractional phase-locked loop ultimately outputs a reference signal that is phase-locked with the rotation frequency of the shaft.

[0047] Furthermore, the lock-in amplifier includes a multiplier, a low-pass filter, and a quadrature demodulation module;

[0048] The multiplier multiplies the acoustic signal with the reference signal and its quadrature signal respectively to generate in-phase and quadrature components;

[0049] The low-pass filter extracts the DC component to obtain the same-direction component. and orthogonal components ;

[0050] The quadrature demodulation module calculates the synthesized amplitude using the in-phase and quadrature components. and phase difference :

[0051]

[0052]

[0053] Based on the synthesized amplitude and preset reference amplitude and reference phase The vibration amplitude and phase difference are calculated using the following formula:

[0054]

[0055]

[0056] in, Indicates the vibration amplitude. This indicates the phase difference.

[0057] The advantages and beneficial effects of this invention are as follows:

[0058] The present invention relates to a bearing fault acoustic detection method and system based on fractional harmonic phase-locked loop (FLL). FLL generates a reference signal synchronized with the bearing fault characteristic frequency. A lock-in amplifier is used to perform phase-sensitive detection on the acoustic signal and the reference signal to obtain the vibration amplitude and phase information corresponding to the bearing fault characteristic frequency. Based on the comparison of the vibration amplitude with a preset baseline, harmonic consistency, and phase stability, the bearing fault type and severity are determined. The present invention, through the coordinated design of FLL and a lock-in amplifier, avoids interference from background noise and vibrations from other equipment in industrial environments, thereby improving the accuracy of bearing fault detection. Attached Figure Description

[0059] Figure 1 This is a flowchart of the method in an embodiment of the present invention.

[0060] Figure 2a This is a front perspective view of the bearing geometry in an embodiment of the present invention.

[0061] Figure 2b This is a side cross-sectional view of the bearing geometry in an embodiment of the present invention.

[0062] Figure 3 This is a schematic diagram of the fractional PLL reference signal generation principle in an embodiment of the present invention.

[0063] Figure 4 This is a block diagram of the fractional harmonic phase-locked amplifier system (including PLL, multiplier, and low-pass filter) in an embodiment of the present invention.

[0064] Figure 5 This is the overall architecture diagram of the system in this embodiment of the invention (including PLL, LIA and signal processing flow). Detailed Implementation

[0065] The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0066] like Figure 1 As shown, this invention relates to an acoustic detection method for bearing faults based on fractional harmonic lock-in amplification. In practical applications, it is necessary to closely integrate the structure and characteristics of the bearing. The following will be described in detail with reference to the accompanying drawings.

[0067] like Figure 2a , Figure 2b As shown, a bearing mainly consists of rolling elements, an inner ring, an outer ring, and a cage (not specifically marked in the diagram but actually present). Among these, The pitch circle diameter is a key dimensional parameter in the bearing structure, which directly affects the motion trajectory and stress of the rolling elements. The diameter of the rolling element is the bearing's diameter. The rolling element plays a role in transmitting load and ensuring smooth rotation during the operation of the bearing. Its diameter is closely related to the bearing's load-bearing capacity and operational stability. The number of rolling elements affects the bearing's load-carrying capacity and the calculation of its failure characteristic frequency. This is the angular frequency of the motor, from which the rotational frequency of the motor shaft can be calculated. The calculation formula is as follows:

[0068]

[0069] in, It is the shaft rotation frequency of the motor. The shaft rotation frequency is the base frequency for subsequent calculation of bearing fault characteristic frequencies. All characteristic frequencies related to bearing faults are related to the shaft frequency. The contact angle is the angle between the bearing and the outer raceway. The size of the contact angle changes the contact state and force distribution between the rolling elements and the raceway, which has a significant impact on the bearing's performance and failure frequency.

[0070] In the actual implementation of the bearing fault detection of this invention, the first step is to obtain the accurate parameters of the bearing being used, i.e. , , and For example, a bearing with model number 6203ZZ, its... =8, =40mm, =0.79375mm, which can be assumed in some cases. = (When the actual contact angle is close to hour, (It has little impact on the calculation results). Based on these parameters, combined with the formula:

[0071]

[0072]

[0073]

[0074]

[0075] in, These are the bearing inner ring fault characteristic frequency (in this embodiment, the inner ring ball passage frequency), the bearing outer ring fault characteristic frequency (in this embodiment, the outer ring ball passage frequency), the ball spin frequency, and the cage passage frequency; and the terms within the curly braces can be simplified to dimensionless constants. (in The constants (BPFI, BPFO, BSF, and FTF) are simplified because they depend on the bearing geometry. It's important to note that these fault-related frequencies are the shaft rotational frequencies. The non-integer multiples of the harmonic frequency can then be calculated using the following formula:

[0076]

[0077] These fault frequencies and harmonics are important indicators for determining whether a bearing is faulty.

[0078] Then, the rotational frequency of the motor shaft is monitored in real time using a tachometer. This generates a synchronized square wave or pulse signal, calculated as described above. The frequency of this signal is... (For example, 29.96 Hz). The input signal undergoes initial frequency division by a prescaler to match the operating range of subsequent circuits. Then, the calculated... Input into the fractional PLL frequency synthesizer, such as Figure 3As shown, a fractional-N PLL is a frequency synthesizer capable of generating an output frequency that is not an integer multiple of the input frequency. Its core function is to generate a reference signal synchronized with bearing fault characteristic frequencies (such as BPFI, BPFO, etc.) based on the motor shaft speed (obtained via a tachometer). The detailed workflow is as follows: A phase frequency detector (PFD) compares the phase of the input reference signal with the feedback signal from the voltage-controlled oscillator (VCO) after it has been divided by a frequency divider. It outputs an error signal (usually a pulse signal) proportional to the phase difference, which drives a charge pump to generate a voltage signal. The voltage signal output by the charge pump passes through a low-pass filter (LPF) to filter out high-frequency noise, generating a smooth DC control voltage. This voltage is used to adjust the output frequency of the VCO, ensuring that the feedback signal is phase-locked with the input signal. The VCO generates a high-frequency signal based on the control voltage output from the loop filter, and its output frequency... It has a linear relationship with the control voltage; for example, if the target frequency is the bearing failure frequency. Then the VCO needs to output ,in This is the division factor; then the frequency divider dynamically divides the VCO output signal, with the division ratio being a non-integer (e.g., 4.793 corresponds to BPFI). The frequency divider achieves fractional frequency division by alternating the averaging effect of integer division values ​​(such as 4 and 5). For example, a division ratio of 4.793 can be achieved by dividing the frequency by 4 793 times and by 5 207 times out of every 1000 cycles. The divided signal is fed back to the PFD to form a closed-loop control, ensuring that the output frequency is precisely locked to the target fractional harmonic. Finally, the fractional PLL outputs a frequency that is consistent with the motor shaft frequency. Phase locked and frequency is sinusoidal reference signal ( Fault type (For harmonic orders), the signal is split into two paths: one path is directly input to the multiplier, and the other path passes through... After phase shift, an orthogonal reference signal is generated.

[0079] Lock-in amplifiers (LIAs) extract weak signals at specific frequencies from noise using phase-sensitive detection (PSD). Their core modules include multipliers, low-pass filters, and quadrature demodulation, such as... Figure 4 As shown, the detailed process is as follows:

[0080] First, the acoustic signals collected by the microphone Includes bearing vibration components (such as) and environmental noise The acoustic signal is transmitted through channels 1 and 2. The reference signals generated by the fractional PLL are fed into multiplier 1 and multiplier 2 respectively. and its orthogonal signals Input the signal into LIA, and then input the signal into channel 1. With reference signal Multiply to get Input signal in channel 2 Orthogonal reference signal Multiply to get The multiplication operation converts the target frequency component into a DC component and a high-frequency component. The outputs of the two multipliers then pass through low-pass filters (such as 3rd-order Butterworth filters), with cutoff frequencies much lower than... (For example, 0.6Hz), high-frequency components are filtered out, retaining only the DC component. The calculation formula is as follows:

[0081]

[0082]

[0083] Next, the composite amplitude is calculated by transforming from rectangular coordinates to polar coordinates. and phase difference The calculation formula is as follows:

[0084]

[0085]

[0086] From the composite amplitude and known reference amplitude The amplitude of the vibration signal can be deduced. The calculation formula is as follows:

[0087]

[0088] The time constant of a low-pass filter (For example, 1 second) determines the smoothness of the output response and the noise suppression capability. The amplitude and phase fluctuations are minimal in steady state, indicating that the target frequency has been successfully locked.

[0089] The following section explains phase-sensitive detection (PSD) and its specific process:

[0090] Phase-sensitive detection (PSD) is a key technology in lock-in amplifiers (LIA), primarily used to accurately extract the amplitude and phase information of a specific frequency signal from a noisy signal. In bearing fault detection, the sound signal generated by the motor contains multiple frequency components and noise. PSD can separate the weak signal related to the characteristic frequency of the bearing fault from the complex background. By comparing its phase with a reference signal, the amplitude and phase of this signal can be obtained, providing crucial information for determining whether a bearing fault exists. First, the signal to be detected... It can be represented as:

[0091]

[0092] in It is the signal amplitude. It is the signal frequency. It is the signal phase. It's noise. The reference signal can be represented as:

[0093]

[0094] in It is the amplitude of the reference signal. It is the reference signal frequency. It is the phase of the reference signal. It is an auxiliary variable. In actual testing, it will... This ensures that the target signal can be detected. The signal detected by phase-sensitive sensing is then described by the following formula:

[0095]

[0096] By expanding and simplifying using trigonometric identities, we can obtain:

[0097]

[0098] At this point, the signal contains low-frequency terms (related to the signal phase difference) and high-frequency terms (as well as the product of noise and the reference signal). To obtain low-frequency information related to the signal phase, [the following steps are taken]. When using a low-pass filter, its cutoff frequency is... To meet After low-pass filtering, the high-frequency terms are effectively removed, and the new signal mathematical expression is as follows:

[0099]

[0100] This is a DC level signal related to the phase difference between the signal and the reference signal. To solve for the signal amplitude... and phase Introducing auxiliary variables By setting and We obtain two orthogonal signals, with the following specific expressions:

[0101]

[0102]

[0103] Will and Treating them as two components of a vector, the magnitude of the composite vector can be obtained using vector calculation methods:

[0104]

[0105] and phase difference :

[0106]

[0107] The amplitude of the signal vibration can then be derived:

[0108]

[0109] Through these calculations, the amplitude and phase information of the signal to be detected are successfully extracted from the noisy signal, which is then used for subsequent bearing fault diagnosis. This is the process of Phase Sensitive Detection (PSD) technology.

[0110] Finally, statistical analysis was performed on the large amount of data collected under normal conditions. Statistical measures such as the average amplitude and standard deviation were calculated, and these were used as the basis for setting amplitude threshold ranges. For example, the average amplitude at a certain fault characteristic frequency was calculated as follows: The standard deviation is The normal range of amplitude may be set to , The value of is determined based on the actual situation and the required detection accuracy, and is generally 2 or 3. For the phase, its normal fluctuation range is determined through multiple measurements, for example... The phase and amplitude calculated by the aforementioned lock-in amplifier (LIA) using phase-sensitive detection (PSD) technology are compared with the preset amplitude threshold range and phase threshold range. If the amplitude increases significantly and the phase is abnormal, and both exceed their respective preset thresholds, it is determined that the bearing is faulty. If both the amplitude and phase are within the set normal range, it can be determined that the bearing is in normal operating condition.

[0111] like Figure 5 As shown, the bearing fault acoustic detection system based on fractional harmonic phase-locked loop amplification includes a fractional phase-locked loop, a phase-locked amplifier, and a fault diagnosis module.

[0112] The fractional phase-locked loop generates a reference signal synchronized with the bearing fault characteristic frequency based on the acquired shaft rotation frequency; wherein, the bearing fault characteristic frequency is calculated based on the bearing's geometric parameters and is a non-integer multiple of the shaft rotation frequency;

[0113] The fractional phase-locked loop (PLL) includes a phase frequency detector, a charge pump, a low-pass filter, and a fractional frequency divider. The phase frequency detector compares the acquired shaft rotation frequency with the feedback signal in phase, outputting an error signal proportional to the phase difference to drive the charge pump to generate a voltage signal. The low-pass filter filters the voltage signal, removing high-frequency noise and generating a smooth DC control voltage. This voltage is used to adjust the output frequency of the voltage-controlled oscillator (VCO), ensuring that the feedback signal is phase-locked with the input signal. The VCO generates a high-frequency signal based on the DC control voltage, and its output frequency is linearly related to the control voltage. The fractional frequency divider dynamically divides the high-frequency signal to obtain the feedback signal, forming a closed-loop control that ensures the output frequency is precisely locked to the target fractional harmonic, so that the PLL ultimately outputs a reference signal that is phase-locked with the shaft rotation frequency.

[0114] The lock-in amplifier performs phase-sensitive detection on the acoustic signals of the shaft and bearing and the corresponding reference signal; the acoustic signal and the reference signal are orthogonally demodulated to generate in-phase and quadrature components, and then the DC component is extracted by low-pass filtering. After demodulation by the lock-in amplifier, the vibration amplitude and phase information corresponding to the bearing fault characteristic frequency are obtained.

[0115] The lock-in amplifier includes a multiplier, a low-pass filter, and a quadrature demodulation module. The multiplier multiplies the acoustic signal with the reference signal and its quadrature signal respectively to generate in-phase and quadrature components. The low-pass filter extracts the DC component to obtain the in-phase component. and orthogonal components The quadrature demodulation module calculates the synthesized amplitude using both in-phase and quadrature components. and phase difference :

[0116]

[0117]

[0118] Based on the synthesized amplitude and preset reference amplitude and reference phase The vibration amplitude and phase difference are calculated using the following formula:

[0119]

[0120]

[0121] in, Indicates the vibration amplitude. This indicates the phase difference.

[0122] The fault diagnosis module determines the type and severity of bearing faults based on the comparison of the vibration amplitude with a preset baseline, harmonic consistency, and phase stability.

[0123] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An acoustic detection method for bearing faults based on fractional harmonic lock-in amplification, characterized in that... Includes the following steps: Acquire acoustic signals from the mating shaft and bearings; Obtain the shaft rotation frequency and synchronize it to the fractional phase-locked loop; The bearing fault characteristic frequency is calculated based on the bearing's geometric parameters, and the bearing fault characteristic frequency is a non-integer multiple of the shaft rotation frequency; The fractional phase-locked loop (PLL) generates a reference signal synchronized with the bearing fault characteristic frequency. The PLL acquires the shaft rotation frequency and compares the phase of the shaft rotation frequency with the feedback signal from the voltage-controlled oscillator (VCO) after frequency division using a phase frequency detector. It outputs an error signal proportional to the phase difference to drive a charge pump to generate a voltage signal. The voltage signal is filtered by a low-pass filter to remove high-frequency noise and generate a DC control voltage. The VCO generates a high-frequency signal based on the DC control voltage, and its output frequency is linearly related to the control voltage. The high-frequency signal is dynamically divided by a fractional frequency divider, and the divided signal is fed back to the VCO to form a closed-loop control. Finally, the PLL outputs a reference signal that is phase-locked with the shaft rotation frequency. Phase-sensitive detection of the acoustic signal and the reference signal is performed using a lock-in amplifier; The acoustic signal and the reference signal are orthogonally demodulated to generate in-phase and quadrature components. The DC component is then extracted by low-pass filtering. After orthogonal demodulation, the vibration amplitude and phase information corresponding to the bearing fault characteristic frequency are obtained. Based on the comparison of the vibration amplitude with the preset baseline, harmonic consistency, and phase stability, the bearing fault type and severity are determined.

2. The bearing fault acoustic detection method based on fractional harmonic lock-in amplification according to claim 1, characterized in that: The dynamic frequency division has a non-integer division ratio. The frequency divider achieves fractional frequency division by alternately switching the average effect of integer frequency division values.

3. The bearing fault acoustic detection method based on fractional harmonic lock-in amplification according to claim 1, characterized in that: The bearing failure characteristic frequency is based on the acquired shaft rotation frequency and the number of bearing balls. Ball diameter Pitch circle diameter and contact angle It can be obtained through the following formula: in, This indicates the characteristic frequency of bearing inner ring failure. The characteristic frequency of bearing outer ring failure is represented by m. BPFI The non-integer multiple coefficient representing the characteristic frequency of bearing inner ring failure, m BPFO A non-integer multiple coefficient representing the characteristic frequency of bearing outer ring failure. Indicates the rotational frequency of the shaft. Represents angular frequency. This indicates the contact angle, which is the angle between the bearing and the outer raceway.

4. The bearing fault acoustic detection method based on fractional harmonic lock-in amplification according to claim 1, characterized in that: The lock-in amplifier multiplies the acoustic signal with the reference signal and its quadrature signal respectively to generate in-phase and quadrature components. The DC component is then extracted using a low-pass filter to obtain the in-phase component of the DC component. Orthogonal components of DC component ; The composite amplitude is calculated using the same-direction component and the quadrature component of the DC component. and phase difference : Based on the synthesized amplitude and preset reference amplitude and reference phase The vibration amplitude and phase difference are calculated using the following formula: in, Indicates the vibration amplitude. This indicates the phase difference.

5. The bearing fault acoustic detection method based on fractional harmonic lock-in amplification according to claim 1, characterized in that: The fractional phase-locked loop integrated frequency synthesizer is used to generate multi-order harmonic reference signals as an extension of the reference signal, which are used to detect different fault characteristic frequencies that may exist in the bearing. The calculation formula is as follows: in, This represents the multi-order harmonic reference signal. A preset positive integer representing the order of the harmonics. Non-integer multiples of the characteristic frequency of bearing failure are represented by coefficients. Indicates the rotational frequency of the shaft; The acoustic signal and the multi-order harmonic reference signal are orthogonally demodulated to generate multi-order in-phase components and multi-order quadrature components. Then, the multi-order DC components are extracted by low-pass filtering, and the multi-order harmonic vibration amplitude and multi-order phase information corresponding to the bearing fault characteristic frequency are calculated.

6. The bearing fault acoustic detection method based on fractional harmonic lock-in amplification according to claim 5, characterized in that: The bearing failure was determined as follows: Under fault-free conditions, measure and record the baseline threshold of vibration amplitude at the characteristic frequencies of each bearing failure; acquire the vibration amplitude of the target bearing in real time. Phase difference And the amplitude of multi-order harmonic vibrations; When the vibration amplitude is measured in real time A fault warning is triggered when the baseline threshold is exceeded. Verify whether the amplitude of multi-order harmonic vibrations corresponding to the characteristic frequency of bearing failure increases synchronously; Calculate phase difference Standard deviation ,like If the signal is less than 0.1 rad, it is confirmed that the signal originates from the target bearing. The fault location is determined based on the characteristic frequency of the abnormal vibration amplitude, and the severity is classified according to the multiple by which the vibration amplitude exceeds the baseline; wherein, the multiple is positively correlated with the severity.

7. An acoustic detection system for bearing faults based on fractional harmonic phase-locked loop amplification, comprising a fractional phase-locked loop, a phase-locked amplifier, and a fault diagnosis module, characterized in that: The fractional phase-locked loop (PLL) generates a reference signal synchronized with the bearing fault characteristic frequency based on the acquired shaft rotation frequency. The bearing fault characteristic frequency is calculated based on the bearing's geometric parameters and is a non-integer multiple of the shaft rotation frequency. The PLL includes a phase frequency detector, a charge pump, a low-pass filter, and a fractional frequency divider. The phase frequency detector compares the acquired shaft rotation frequency with the feedback signal from the voltage-controlled oscillator (VCO) after frequency division, outputting an error signal proportional to the phase difference to drive the charge pump to generate a voltage signal. The low-pass filter filters the voltage signal, removing high-frequency noise and generating a DC control voltage. The VCO generates a high-frequency signal based on the DC control voltage, with its output frequency linearly related to the control voltage. The fractional frequency divider dynamically divides the high-frequency signal, feeding the divided signal back to the VCO to form a closed-loop control, ensuring the PLL ultimately outputs a reference signal phase-locked with the shaft rotation frequency. The lock-in amplifier performs phase-sensitive detection on the acoustic signals of the shaft and bearing and the corresponding reference signals; it performs quadrature demodulation on the acoustic signals and the reference signals to generate in-phase and quadrature components, then extracts the DC component through low-pass filtering, and after demodulation by the lock-in amplifier, obtains the vibration amplitude and phase information corresponding to the bearing fault characteristic frequency; The fault diagnosis module determines the type and severity of bearing faults based on the comparison of the vibration amplitude with a preset baseline, harmonic consistency, and phase stability.

8. The bearing fault acoustic detection system based on fractional harmonic lock-in amplification according to claim 7, characterized in that: The lock-in amplifier includes a multiplier, a low-pass filter, and a quadrature demodulation module; The multiplier multiplies the acoustic signal with the reference signal and its quadrature signal respectively to generate in-phase and quadrature components; The low-pass filter extracts the DC component to obtain the same-direction component. and orthogonal components ; The quadrature demodulation module calculates the synthesized amplitude using the in-phase and quadrature components. and phase difference : Based on the synthesized amplitude and preset reference amplitude and reference phase The vibration amplitude and phase difference are calculated using the following formula: in, Indicates the vibration amplitude. This indicates the phase difference.

Images