Motor fault detection method and device

Abstract

The invention discloses a motor fault detection method and device, and relates to the technical field of fault detection, and the method comprises the steps: obtaining an initial acceleration time domain signal collected by a three-axis vibration sensor disposed on a motor, a background vibration signal collected by a reference sensor disposed on a motor pedestal, and a motor three-phase current; performing sliding direct current removal processing on the initial acceleration time-domain signal to obtain a reference acceleration time-domain signal; performing current notch filtering on the reference acceleration time-domain signal based on the three-phase current of the motor to obtain a candidate acceleration time-domain signal; performing environment vibration coherent filtering on the candidate acceleration time-domain signal based on the background vibration signal to obtain a target acceleration time-domain signal; performing feature enhancement on the target acceleration time-domain signal to obtain an enhanced target acceleration time-domain signal; and fault detection is carried out on the motor according to the enhanced target acceleration time domain signal, and the accuracy of motor fault detection is effectively improved through multi-stage noise reduction processing.

Description

Technical Field

[0001] This application relates to the field of fault detection technology, and in particular to a method and apparatus for detecting motor faults. Background Technology

[0002] In the traditional field of motor fault detection, manual inspection and alarm systems based on simple thresholds are commonly relied upon. These methods often suffer from low detection efficiency, frequent false alarms and missed alarms, and are unable to meet the high requirements of modern industrial production for motor operational stability and safety. Specifically, manual inspection is not only time-consuming and labor-intensive, but also limited by the experience and condition of the inspectors, making real-time and comprehensive fault monitoring difficult. Alarm systems based on simple thresholds, lacking in-depth analysis and understanding of the motor's operating status, often fail to accurately distinguish between normal fluctuations and genuine faults, leading to frequent false alarms and missed alarms. Summary of the Invention

[0003] The main objective of this application is to provide a method and apparatus for detecting motor faults, which aims to solve the technical problems of low detection efficiency and low accuracy in traditional motor fault detection methods.

[0004] To achieve the above objectives, this application proposes a motor fault detection method, which includes: The initial acceleration time-domain signal collected by the triaxial vibration sensor mounted on the motor, the background vibration signal collected by the reference sensor mounted on the motor base, and the three-phase current of the motor are acquired, wherein the triaxial vibration sensor is in rigid contact with the motor housing; The initial acceleration time-domain signal is subjected to a sliding DC removal process to obtain the reference acceleration time-domain signal; Based on the three-phase current of the motor, the reference acceleration time-domain signal is subjected to current notch filtering to obtain the candidate acceleration time-domain signal; Based on the background vibration signal, the candidate acceleration time-domain signal is subjected to environmental vibration coherent filtering to obtain the target acceleration time-domain signal; The target acceleration time-domain signal is enhanced to obtain the enhanced target acceleration time-domain signal. The motor is used for fault detection based on the enhanced target acceleration time-domain signal.

[0005] In one embodiment, the step of performing current notch filtering on the reference acceleration time-domain signal based on the three-phase current of the motor to obtain the candidate acceleration time-domain signal includes: The sampling frequency, notch depth control parameters, and time delay of the triaxial vibration sensor are obtained. Extract the characteristic frequencies of the current ripple of the three-phase current of the motor; Based on the sampling frequency, the notch depth control parameters, the time delay, and the current ripple characteristic frequency, the reference acceleration time-domain signal is subjected to current notch filtering to obtain the candidate acceleration time-domain signal.

[0006] In one embodiment, the step of performing current notch filtering on the reference acceleration time-domain signal based on the sampling frequency, the notch depth control parameter, the time delay, and the current ripple characteristic frequency to obtain a candidate acceleration time-domain signal includes: The discrete-time transfer function of the adaptive notch filter is constructed based on the sampling frequency, the notch depth control parameters, the time delay, and the current ripple characteristic frequency. The reference acceleration time-domain signal is subjected to current notch filtering based on the discrete-time transfer function to obtain the candidate acceleration time-domain signal.

[0007] In one embodiment, the step of performing a sliding DC removal process on the initial acceleration time-domain signal to obtain a reference acceleration time-domain signal includes: Obtain the sampling frequency and window number of the triaxial vibration sensor; The window duration is calculated based on the sampling frequency and the number of window points. Based on the window duration, the initial acceleration time-domain signal is slid in real time to obtain the sliding acceleration time-domain signal; Calculate the average value of the sliding acceleration time-domain signal within the calculation window; The initial acceleration time-domain signal is de-DC processed based on the average value of the signal to obtain the reference acceleration time-domain signal.

[0008] In one embodiment, the step of performing environmental vibration coherent filtering on the candidate acceleration time-domain signal based on the background vibration signal to obtain the target acceleration time-domain signal includes: Calculate the autopower spectrum of the background vibration signal and the autopower spectrum of the candidate acceleration time-domain signal at a specific frequency; Calculate the cross power spectrum of the background vibration signal and the candidate acceleration time-domain signal at a specific frequency; The candidate acceleration time-domain signal is subjected to environmental vibration coherent filtering based on the cross power spectrum, the autopower spectrum of the background vibration signal, and the autopower spectrum of the candidate acceleration time-domain signal to obtain the target acceleration time-domain signal.

[0009] In one embodiment, the step of performing environmental vibration coherent filtering on the candidate acceleration time-domain signal based on the cross power spectrum, the autopower spectrum of the background vibration signal, and the autopower spectrum of the candidate acceleration time-domain signal to obtain the target acceleration time-domain signal includes: The linear coherence value between the background vibration signal and the candidate acceleration time-domain signal is calculated based on the cross power spectrum, the autopower spectrum of the background vibration signal, and the autopower spectrum of the candidate acceleration time-domain signal. When the linear coherence value is greater than a preset coherence threshold, the candidate acceleration time-domain signal at the specific frequency is eliminated, and the eliminated candidate acceleration time-domain signal is used as the target acceleration time-domain signal.

[0010] In one embodiment, the step of fault detection of the motor based on the enhanced target acceleration time-domain signal includes: The envelope spectrum is obtained from the enhanced target acceleration time-domain signal; Obtain the rated speed of the motor rotor; Determine the critical speed value and target speed range based on the rated speed; In the envelope spectrum, the target amplitude point is determined by searching within the target speed range starting from the speed critical value. The actual rotational speed of the motor rotor is determined based on the target amplitude point; The motor is used for fault detection based on the actual rotational speed and the envelope spectrum.

[0011] In one embodiment, the step of fault detection of the motor based on the actual rotational speed and the envelope spectrum includes: Obtain bearing structural parameters; The theoretical fault characteristic frequency is calculated based on the actual rotational speed and the bearing structural parameters. Detect whether there is a frequency peak in the envelope spectrum that is consistent with the theoretical fault characteristic frequency; When a frequency peak consistent with the theoretical fault characteristic frequency is detected in the envelope spectrum, it is determined that the motor bearing has a fault.

[0012] In one embodiment, the step of performing feature enhancement on the target acceleration time-domain signal to obtain the enhanced target acceleration time-domain signal includes: The target acceleration time-domain signal is transformed to obtain the analytic signal of the target acceleration time-domain signal; Extract the envelope signal based on the parsed signal; Power spectrum analysis is performed on the envelope signal to obtain the envelope spectrum; The enhanced target acceleration time-domain signal is obtained based on the envelope spectrum.

[0013] Furthermore, to achieve the above objectives, this application also proposes a motor fault detection device, which includes: The acquisition module is used to acquire the initial acceleration time-domain signal collected by the triaxial vibration sensor installed on the motor, the background vibration signal collected by the reference sensor installed on the motor base, and the three-phase current of the motor, wherein the triaxial vibration sensor is in rigid contact with the motor housing; The processing module is used to perform a sliding DC removal process on the initial acceleration time-domain signal to obtain a reference acceleration time-domain signal; The processing module is further configured to perform current notch filtering on the reference acceleration time-domain signal based on the three-phase current of the motor to obtain candidate acceleration time-domain signals; The filtering module is used to perform environmental vibration coherent filtering on the candidate acceleration time-domain signal based on the background vibration signal to obtain the target acceleration time-domain signal; The filtering module is also used to perform feature enhancement on the target acceleration time-domain signal to obtain an enhanced target acceleration time-domain signal; The fault detection module is used to detect faults in the motor based on the enhanced target acceleration time-domain signal.

[0014] In addition, to achieve the above objectives, this application also proposes a motor fault detection device, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the motor fault detection method as described above.

[0015] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the motor fault detection method described above.

[0016] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the motor fault detection method described above.

[0017] The one or more technical solutions proposed in this application have at least the following technical effects: Through multi-level signal noise reduction processing, the accuracy of motor fault detection is effectively improved. First, by utilizing the collaborative acquisition of a triaxial vibration sensor and a reference sensor, differentiated monitoring of motor body vibration and environmental vibration is achieved. Second, DC offset interference in the initial signal is eliminated through sliding de-DC processing, and periodic noise generated by the motor's electromagnetic field is specifically filtered out using current notch filtering technology. Furthermore, an environmental vibration coherent filtering algorithm is employed to effectively suppress signal interference caused by non-motor body vibrations. Finally, feature enhancement processing highlights the fault characteristic frequency band, enabling time-domain signal-based fault discrimination to have a higher signal-to-noise ratio and detection sensitivity. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating an embodiment of the motor fault detection method of this application. Figure 2 A schematic diagram of bearing fault characteristic frequencies calculated based on actual rotational speed 1X frequency, provided as an embodiment of the motor fault detection method of this application; Figure 3 This is a flowchart illustrating Embodiment 2 of the motor fault detection method of this application; Figure 4 This is a flowchart illustrating Embodiment 3 of the motor fault detection method of this application; Figure 5 A simplified flowchart is provided for one embodiment of the motor fault detection method of this application; Figure 6 This is a schematic diagram of the module structure of the motor fault detection device according to an embodiment of this application; Figure 7 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the motor fault detection method in this application embodiment.

[0021] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0022] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0023] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0024] The main solution of this application embodiment is as follows: Acquire the initial acceleration time-domain signal collected by a triaxial vibration sensor mounted on the motor, the background vibration signal collected by a reference sensor mounted on the motor base, and the three-phase current of the motor, wherein the triaxial vibration sensor is in rigid contact with the motor housing; perform a sliding DC-free process on the initial acceleration time-domain signal to obtain a reference acceleration time-domain signal; perform current notch filtering on the reference acceleration time-domain signal based on the motor three-phase current to obtain candidate acceleration time-domain signals; perform environmental vibration coherent filtering on the candidate acceleration time-domain signals based on the background vibration signal to obtain a target acceleration time-domain signal; perform feature enhancement on the target acceleration time-domain signal to obtain an enhanced target acceleration time-domain signal; and perform fault detection on the motor based on the enhanced target acceleration time-domain signal.

[0025] The existing technology for motor vibration detection faces three major noise interference problems: (1) Harmonics of motor winding current are coupled to the vibration sensor through electromagnetic field, resulting in non-mechanical harmonics in the spectrum. (2) When equipment groups are running in tandem, adjacent mechanical vibrations are transmitted to the motor under test through the mounting base, and the time domain waveform exhibits aliasing. (3) Low amplitude vibration signals (<0.05g) are submerged by sensor circuit noise, and traditional filtering methods result in the loss of characteristic frequencies.

[0026] This application provides a solution for analytically locating vibration sources using a dual-channel measurement system comprised of a triaxial vibration sensor and a reference sensor. Specifically, a rigidly contact-mounted triaxial accelerometer acquires the vibration signal of the motor body, while a reference sensor mounted on the motor base simultaneously obtains the environmental vibration baseline. The vibration source is separated through time-frequency domain coherent analysis. At the signal processing level, the original vibration signal is first subjected to sliding window deDC processing to eliminate baseline offset caused by sensor temperature drift. Then, adaptive notch filtering based on motor current ripple characteristics is used to accurately suppress electromagnetic noise interference. Finally, environmental vibration coherent filtering technology is employed to remove vibration interference transmitted from adjacent devices. Addressing the challenge of extracting weak fault features, this solution constructs an envelope modulation model through analytical signal transformation and combines it with power spectral density analysis to achieve adaptive enhancement of the fault feature frequency band.

[0027] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device or motor fault detection device capable of performing the above functions. The following description uses a motor fault detection device as an example to illustrate this embodiment and the subsequent embodiments.

[0028] Based on this, embodiments of this application provide a method for detecting motor faults, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the motor fault detection method of this application.

[0029] In this embodiment, the motor fault detection method includes steps S10 to S60: Step S10: Acquire the initial acceleration time-domain signal collected by the triaxial vibration sensor installed on the motor, the background vibration signal collected by the reference sensor installed on the motor base, and the three-phase current of the motor, wherein the triaxial vibration sensor is in rigid contact with the motor housing.

[0030] It should be noted that in this embodiment, the triaxial MEMS vibration sensor (MPU6050) is mechanically coupled to the motor housing via a rigid connector. The installation resonant frequency is >10kHz to avoid the power frequency harmonic range and ensure high-fidelity transmission of the vibration signal. A high-sensitivity piezoelectric accelerometer is used as the reference sensor, and its installation position is symmetrically arranged at the same height as the main sensor to effectively eliminate the influence of gravitational acceleration components on the measurement results. The three-phase current of the motor is synchronously acquired using a Hall current sensor.

[0031] In practical implementation, for example, if it is necessary to detect bearing failure in a variable frequency motor, a triaxial vibration sensor can be installed on the bearing housing at the motor drive end, the sampling frequency can be set to 12.8kHz, and the motor can be accelerated from 0 to 3000rpm (the carrier frequency of the frequency converter is 4kHz). At the same time, the initial acceleration time domain signal is collected by the triaxial vibration sensor, the background vibration signal is collected by the reference sensor, and the three-phase current of the motor is collected by the Hall sensor.

[0032] Background vibration signals are vibration signals generated by other equipment in the motor's operating environment besides the motor under test. These signals may be transmitted to the motor under test through the mounting base, interfering with the motor's own vibration signal. By acquiring background vibration signals through a reference sensor, an accurate environmental vibration baseline can be obtained, providing basic data for subsequent coherent environmental vibration filtering.

[0033] Step S20: Perform DC removal processing on the initial acceleration time-domain signal to obtain the reference acceleration time-domain signal.

[0034] In practice, the initial acceleration time-domain signal can be subjected to sliding DC removal processing to eliminate sensor zero-point drift and obtain the acceleration time-domain signal after eliminating zero-point offset, i.e., the reference acceleration time-domain signal.

[0035] In one feasible implementation, step S20 may include steps A11 to A15: Step A11: Obtain the sampling frequency and window number of the triaxial vibration sensor; It should be noted that the sampling frequency of the triaxial vibration sensor can be obtained, for example, the sampling frequency is set to 12.8kHz, and the number of window points is the pseudocode number of points, specifically set to 1024.

[0036] Step A12: Calculate the window duration based on the sampling frequency and the number of window points; In practical implementation, the window parameters, namely the window duration, can be calculated based on the sampling frequency and the number of window points. The window duration T = number of window points / sampling frequency. For example, if the sampling frequency is 12.8kHz, then the window duration T = 80ms, which covers the 50Hz power frequency cycle and effectively suppresses the power frequency DC component.

[0037] Step A13: Slide the initial acceleration time-domain signal in real time based on the window duration to obtain the sliding acceleration time-domain signal; In practice, the initial acceleration time-domain signal acquired continuously can be truncated to a window of length T with the current time t as the endpoint, thereby obtaining the sliding acceleration time-domain signal.

[0038] Step A14: Calculate the average signal value of the sliding acceleration time-domain signal within the window; Specifically, the average signal value within the window can be calculated using the following formula: Signal average value =

[0039] Step A15: Perform DC removal processing on the initial acceleration time-domain signal based on the average value of the signal to obtain the reference acceleration time-domain signal.

[0040] It should be noted that the initial acceleration time-domain signal can be deDCed based on the average signal value. This is done by subtracting the average signal value from the initial acceleration time-domain signal to obtain the deDCed signal, which is the reference acceleration time-domain signal. For example, if the average signal value within the window is 0.03g, then 0.03g is subtracted from each point in the initial acceleration time-domain signal to eliminate baseline offset. Specifically, the reference acceleration time-domain signal is calculated as follows:

[0041] In the above formula, For reference, the time-domain acceleration signal, This is the initial acceleration time-domain signal.

[0042] Step S30: Perform current notch filtering on the reference acceleration time-domain signal based on the three-phase current of the motor to obtain the candidate acceleration time-domain signal.

[0043] In practice, during motor operation, the three-phase current generates an electromagnetic field, which in turn produces non-mechanical harmonic interference in the vibration sensor through electromagnetic coupling. This harmonic interference is superimposed on the reference acceleration time-domain signal, affecting the accurate extraction of fault characteristics.

[0044] The principle of current notch filtering is to utilize the harmonic characteristics of the three-phase current of a motor to construct an adaptive notch filter, which specifically filters out electromagnetic noise in the reference acceleration time-domain signal. Specifically, the three-phase current of the motor is first subjected to spectral analysis to obtain the frequency components and amplitude characteristics of the current harmonics. Then, the parameters of the notch filter, such as the notch frequency and bandwidth, are designed based on these characteristics. Finally, the designed notch filter is applied to the reference acceleration time-domain signal to obtain a candidate acceleration time-domain signal with electromagnetic noise interference removed.

[0045] Step S40: Perform environmental vibration coherent filtering on the candidate acceleration time-domain signal based on the background vibration signal to obtain the target acceleration time-domain signal.

[0046] It should be noted that when equipment groups operate collaboratively, adjacent mechanical vibrations can be transmitted to the motor under test through the mounting base, causing non-motor vibration components to be mixed into the candidate acceleration time-domain signal. These environmental vibration interferences exhibit aliasing characteristics in the time-domain waveform, and if not filtered out specifically, they will directly reduce the accuracy of fault detection.

[0047] The core of coherent filtering for environmental vibration lies in establishing a time-frequency domain correlation model between the main sensor (triaxial vibration sensor) and the reference sensor. Frequency bands in the candidate signal whose coherence with environmental vibration exceeds a threshold are suppressed. The resulting target acceleration time-domain signal exhibits effective attenuation of non-motor vibration components, while retaining motor fault characteristics.

[0048] Step S50: Perform feature enhancement on the target acceleration time-domain signal to obtain the enhanced target acceleration time-domain signal.

[0049] It should be noted that feature enhancement primarily aims to highlight the characteristic frequency bands of motor faults, thereby improving the sensitivity and accuracy of fault detection. In practice, analytic signal transformation technology can be used to convert the time-domain signal into an analytic signal, thereby constructing an envelope spectrum. This envelope spectrum can extract the modulation components in the signal, which are often closely related to the periodic impacts caused by mechanical faults.

[0050] In one feasible implementation, step S50 may include steps A21 to A24: Step A21: Perform signal transformation on the target acceleration time-domain signal to obtain the analytic signal of the target acceleration time-domain signal; It should be noted that the signal enhancement in this embodiment mainly involves envelope demodulation analysis of the target acceleration time-domain signal. Specifically, the target acceleration time-domain signal is first transformed, mainly by performing a Hilbert transform, to obtain the analytical signal of the target acceleration time-domain signal.

[0051] In practical implementation, the analytical signal is represented as follows: z(t) = a(t) + j (t) In the above formula, z(t) is the analytic signal, and a(t) is the target acceleration time-domain signal. (t) is the signal obtained by performing Hilbert transform on the time-domain signal of the target acceleration, that is, the 90° phase-shifted signal of a(t), where j is the imaginary unit.

[0052] Step A22: Extract the envelope signal based on the parsed signal; In practical implementation, the magnitude of the analyzed signal is the envelope signal, which is represented as follows:

[0053] In the above formula, This is the envelope signal.

[0054] Step A23: Perform power spectrum analysis on the envelope signal to obtain the envelope spectrum; It should be noted that a fast Fourier transform can be performed on the envelope signal to analyze the power spectrum of the envelope signal and obtain the envelope spectrum.

[0055] It is understandable that the peak value on the frequency axis of the envelope spectrum corresponds to the bearing fault characteristic frequency.

[0056] Step A24: Obtain the enhanced target acceleration time-domain signal based on the envelope spectrum.

[0057] In practical implementation, the envelope spectrum can be used as the enhanced time-domain signal of the target acceleration.

[0058] Step S60: Perform fault detection on the motor based on the enhanced target acceleration time-domain signal.

[0059] It should be noted that the envelope spectrum can be obtained from the enhanced target acceleration time-domain signal, and the presence of a motor fault can be determined based on the frequency values ​​on the envelope spectrum.

[0060] In some feasible implementations, step S60 may include steps A31 to A36: Step A31: Obtain the envelope spectrum based on the enhanced target acceleration time-domain signal; In practice, the envelope spectrum can clearly reveal the characteristic frequency components of the motor vibration signal, which are closely related to the motor's fault state. Through detailed analysis of the envelope spectrum, it is possible to accurately identify whether a fault exists in the motor and the type of fault.

[0061] Step A32: Obtain the rated speed of the motor rotor; It should be noted that, for example, by performing dynamic tracking detection on the motor rotor, the rated speed f_n of the motor rotor can be obtained.

[0062] Step A33: Determine the critical speed value and target speed range based on the rated speed; In practice, the speed critical value can be calculated based on the rated speed. For example, the speed critical value is 60% of the rated speed, and the target speed range is expressed as [speed critical value, rated speed].

[0063] Step A34: In the envelope spectrum, starting from the speed critical value, search within the target speed range to determine the target amplitude point; Understandably, the maximum amplitude point, i.e. the target amplitude point, can be searched within the target speed range, starting from f_n×60% in the envelope spectrum.

[0064] Step A35: Determine the actual rotational speed of the motor rotor based on the target amplitude point; It should be noted that the actual rotational speed of the motor rotor can be determined based on the frequency value corresponding to the target amplitude point, combined with the rated rotational speed of the motor rotor and the correspondence between rotational speed and frequency. During motor operation, there is a specific correlation between the rotor rotational speed and the frequency of the vibration signal. By analyzing the frequency corresponding to the target amplitude point in the envelope spectrum, the actual rotational speed of the rotor can be deduced.

[0065] Step A36: Perform fault detection on the motor based on the actual rotational speed and the envelope spectrum.

[0066] In practice, the presence of a bearing fault in the motor can be determined based on the actual rotational speed and the envelope spectrum.

[0067] In one feasible implementation, step A36 may include: obtaining bearing structural parameters; calculating theoretical fault characteristic frequencies based on the actual rotational speed and the bearing structural parameters; detecting whether there is a frequency peak in the envelope spectrum that matches the theoretical fault characteristic frequency; and determining that the motor bearing has a fault when a frequency peak in the envelope spectrum that matches the theoretical fault characteristic frequency is detected.

[0068] It should be noted that bearing structural parameters, including the bearing's inner diameter, outer diameter, rolling element diameter, and number of rolling elements, are crucial for accurately calculating the theoretical fault characteristic frequency. Based on the obtained actual rotational speed and bearing structural parameters, the theoretical fault characteristic frequency can be precisely calculated. This frequency is the specific frequency value exhibited in the frequency spectrum by the vibration signal of the motor bearing during normal operation and when a fault occurs, and it is a key basis for determining whether the bearing is faulty.

[0069] For example, bearing structural parameters include the number of rolling elements n, the rolling element diameter d, the bearing pitch circle diameter D, and the rolling element contact angle α. The theoretical failure frequency of the outer ring in the theoretical failure characteristic frequency is calculated as follows:

[0070] Where fr is the actual rotational speed frequency of the rotor. This represents the theoretical fault frequency of the outer ring.

[0071] The theoretical fault frequencies of the inner ring in the theoretical fault characteristic frequencies are calculated as follows:

[0072] in, This represents the theoretical fault frequency of the inner ring.

[0073] Subsequently, the envelope spectrum is carefully examined for frequency peaks consistent with the theoretical fault characteristic frequency. The envelope spectrum clearly displays the various frequency components of the vibration signal and their corresponding amplitudes. By comparing the theoretical fault characteristic frequency with the frequency peaks in the envelope spectrum, it is possible to effectively determine whether the motor bearing has a fault. If, during the detection process, a frequency peak consistent with the theoretical fault characteristic frequency is found in the envelope spectrum, then it can be determined that the motor bearing has a fault. This judgment result is based on in-depth analysis and processing of the vibration signal and has high accuracy and reliability. Furthermore, based on the magnitude and location of the frequency peaks, the severity and type of the fault can be further determined, providing a strong basis for subsequent repair and replacement. Figure 2 As shown, Figure 2This diagram illustrates the characteristic frequencies of bearing faults calculated based on the actual rotational speed multiplied by the frequency (BPFO / BPFI) to achieve accurate fault identification in noisy environments. For example, the fundamental frequency of 9.38 Hz (corresponding to an actual rotational speed of 562.8 rpm) is identified in the envelope spectrum. By calculating the bearing outer ring fault frequency: 9.38 × 6.759 = 63.37 Hz, the characteristic lines of 63.37 Hz, 126.73 Hz (2nd harmonic), and 190.1 Hz (3rd harmonic) are clearly visible in the spectrum, indicating damage to the outer ring of the motor. For instance, the equally spaced spectral lines (Δf ≈ 63.3 Hz) conform to the modulation pattern of outer ring faults, and the amplitude of its 4th harmonic is 12 times that of the background noise (signal-to-noise ratio > 2 dB). Moreover, the sideband spacing equals the rotor fundamental frequency (9.38 Hz), thus verifying the coupling relationship of the fault source.

[0074] This embodiment provides a motor fault detection method that uses a dual-channel measurement system consisting of a triaxial vibration sensor and a reference sensor to achieve analytical localization of the vibration source. Specifically, a rigidly contact-mounted triaxial accelerometer is used to collect the vibration signal of the motor body, while an environmental vibration baseline is simultaneously obtained through a reference sensor mounted on the motor base. The vibration source is separated by time-frequency domain coherent analysis. At the signal processing level, the original vibration signal is first subjected to sliding window DC removal processing to eliminate baseline offset caused by sensor temperature drift; then, adaptive notch filtering based on motor current ripple characteristics is used to accurately suppress electromagnetic noise interference; finally, environmental vibration coherent filtering technology is used to eliminate vibration interference transmitted from adjacent devices. To address the challenge of extracting weak fault features, this solution constructs an envelope modulation model through analytical signal transformation and combines it with power spectral density analysis to achieve adaptive enhancement of the fault feature frequency band.

[0075] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in Embodiment 1 above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 3 Step S30 includes steps S301 to S303: Step S301: Obtain the sampling frequency, notch depth control parameters, and time delay of the triaxial vibration sensor.

[0076] It should be noted that the sampling frequency of the triaxial vibration sensor is set to 12.8kHz, and the notch depth control parameter r is in the range of (0,1). For example, it is initially set to the fault value of 0.95. The closer r is to 1, the narrower the notch bandwidth and the greater the attenuation depth of the current ripple characteristic frequency.

[0077] Time delay z 1 is the unit delay operator in discrete time, representing the time delay of the signal.

[0078] Step S302: Extract the current ripple characteristic frequency of the three-phase current of the motor.

[0079] In practical implementation, the harmonic component frequencies extracted from the three-phase current of the motor can be synchronously acquired, namely the current ripple characteristic frequencies, such as winding current harmonics and inverter switching frequencies.

[0080] Step S303: Perform current notch filtering on the reference acceleration time-domain signal based on the sampling frequency, the notch depth control parameter, the time delay, and the current ripple characteristic frequency to obtain the candidate acceleration time-domain signal.

[0081] In practice, the reference acceleration time-domain signal can be filtered by current notch filtering through sampling frequency, notch depth control parameters, time delay, and current ripple characteristic frequency to suppress electromagnetic noise and obtain candidate acceleration time-domain signals.

[0082] In one feasible implementation, step S303 may include steps B11-B12: Step B11: Construct the discrete-time transfer function of the adaptive notch filter based on the sampling frequency, the notch depth control parameters, the time delay, and the current ripple characteristic frequency; It should be noted that the discrete-time transfer function of the adaptive notch filter is expressed as follows:

[0083] In the above formula, The characteristic frequency of the current ripple. Sampling frequency, The time delay is denoted by r, which is the notch depth control parameter and can be determined based on the extracted data. Dynamically adjust filter parameters, by Item precisely aligned with target frequency This ensures that the center frequency of the notch filter is perfectly matched with the frequency of the electromagnetic noise.

[0084] Step B12: Perform current notch filtering on the reference acceleration time-domain signal according to the discrete-time transfer function to obtain the candidate acceleration time-domain signal.

[0085] In practical implementation, the reference acceleration time-domain signal can be substituted into the discrete-time transfer function, and real-time time-domain processing can be achieved through the difference equation to obtain the candidate acceleration time-domain signal, as shown below: y(n)=x(n) 2cos(2πfc / fs)x(n 1)+x(n 2)+2rcos(2πfc / fs)y(n 1) r2y(n 2) Where x(n) is the reference acceleration time-domain signal, and y(n) is the signal after current notch filtering, i.e., the candidate acceleration time-domain signal.

[0086] This embodiment acquires the sampling frequency, notch depth control parameters, and time delay of the triaxial vibration sensor; extracts the current ripple characteristic frequency of the three-phase current of the motor; and performs current notch filtering on the reference acceleration time-domain signal based on the sampling frequency, the notch depth control parameters, the time delay, and the current ripple characteristic frequency to obtain candidate acceleration time-domain signals. This preserves the true characteristics of the motor's mechanical vibration while effectively suppressing electromagnetic noise interference, laying a solid foundation for subsequent accurate fault feature extraction.

[0087] Based on the first embodiment of this application, in the third embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 4 Step S40 includes steps S401 to S403: Step S401: Calculate the autopower spectrum of the background vibration signal and the autopower spectrum of the candidate acceleration time-domain signal at a specific frequency.

[0088] It should be noted that the selection of a specific frequency is usually based on the range of vibration frequencies that may occur during motor operation and the preliminary analysis of the motor's vibration characteristics. Calculating the autopower spectrum of the background vibration signal can clearly reveal the energy distribution of environmental vibration in the frequency domain, which helps to accurately separate the effective components in the motor's own vibration signal. Calculating the autopower spectrum of the candidate acceleration time-domain signal is to obtain the energy characteristics of the motor's mechanical vibration signal in the frequency domain after current notch filtering, so as to further analyze its correlation with fault characteristics.

[0089] In practical implementation, the Fast Fourier Transform (FFT) algorithm can be used to calculate the self-power spectrum. The self-power spectrum of the background vibration signal is usually the pure noise spectrum of the environmental noise, while the self-power spectrum of the candidate acceleration time-domain signal is used to describe the energy distribution of the frequency components of the triaxial vibration sensor itself.

[0090] Step S402: Calculate the cross power spectrum of the background vibration signal and the candidate acceleration time-domain signal at a specific frequency.

[0091] It is understandable that the cross power spectrum between the background vibration signal and the candidate acceleration time-domain signal can be calculated at a specific frequency, and the cross power spectrum reflects the degree of energy coupling between the two at a specific frequency.

[0092] Step S403: Perform environmental vibration coherent filtering on the candidate acceleration time-domain signal based on the cross power spectrum, the autopower spectrum of the background vibration signal, and the autopower spectrum of the candidate acceleration time-domain signal to obtain the target acceleration time-domain signal.

[0093] In practice, the coherence function can be used to measure the linear correlation between the background vibration signal and the candidate acceleration time-domain signal at a specific frequency based on the cross power spectrum, the autopower spectrum of the background vibration signal, and the autopower spectrum of the candidate acceleration time-domain signal, thereby performing environmental vibration coherent filtering on the candidate acceleration time-domain signal.

[0094] In one feasible implementation, step S403 may include steps C11-C12: Step C11: Calculate the linear coherence value between the background vibration signal and the candidate acceleration time-domain signal based on the cross power spectrum, the autopower spectrum of the background vibration signal, and the autopower spectrum of the candidate acceleration time-domain signal; It should be noted that the linear coherence value, i.e. the power spectrum coherence function, is expressed as follows:

[0095] In the above formula, For cross-power spectrum, The power spectrum of the background vibration signal. This represents the autopower spectrum of the candidate acceleration time-domain signal.

[0096] Step C12: When the linear coherence value is greater than the preset coherence threshold, the candidate acceleration time-domain signal at the specific frequency is eliminated, and the eliminated candidate acceleration time-domain signal is used as the target acceleration time-domain signal.

[0097] The linear coherence value ranges from 0 to 1. When the coherence function value is close to 1, it indicates a high degree of linear correlation between the two at that frequency, and significant interference from background vibration on the candidate acceleration time-domain signal. When the coherence function value is close to 0, it indicates a low degree of linear correlation between the two, and less interference from background vibration. By setting an appropriate coherence threshold, when the linear coherence value is greater than the preset threshold, it can be determined that background vibration has significantly interfered with the candidate acceleration time-domain signal at that specific frequency. In this case, the candidate acceleration time-domain signal at that specific frequency can be removed, effectively eliminating the interference components caused by environmental vibration. Using the candidate acceleration time-domain signal after removing interference components as the target acceleration time-domain signal, this signal more purely reflects the mechanical vibration characteristics of the motor body, providing a reliable basis for subsequent accurate extraction of motor fault characteristics, and helping to more accurately determine whether the motor has a fault and the type and severity of the fault. For example, in practical applications, the preset coherence threshold can be reasonably set according to the actual operating environment of the motor and past experience. If it is set to 0.8, when the calculated linear coherence value is greater than 0.8, the rejection operation is performed, thereby obtaining a more accurate target acceleration time domain signal.

[0098] This embodiment calculates the autopower spectrum of the background vibration signal and the autopower spectrum of the candidate acceleration time-domain signal at a specific frequency; it also calculates the cross-power spectrum of the background vibration signal and the candidate acceleration time-domain signal at the specific frequency; and performs environmental vibration coherent filtering on the candidate acceleration time-domain signal based on the cross-power spectrum, the autopower spectrum of the background vibration signal, and the autopower spectrum of the candidate acceleration time-domain signal to obtain the target acceleration time-domain signal. This process effectively eliminates vibration interference transmitted by adjacent equipment, enabling the target acceleration time-domain signal to more accurately reflect the actual vibration of the motor body, providing a high-quality data foundation for subsequent fault feature extraction and analysis.

[0099] For example, to help understand the implementation flow of the motor fault detection method obtained by combining this embodiment with the above embodiment one, please refer to... Figure 5 , Figure 5A simplified flowchart of a motor fault detection method is provided. Specifically, raw signals such as acceleration time-domain signals, environmental vibration signals, and motor current signals are collected and preprocessed in a signal preprocessing layer. The preprocessed acceleration time-domain signal is then input into a noise separation layer for noise suppression. Finally, the noise-suppressed signal is input into a feature enhancement layer for feature enhancement, thereby enabling motor fault detection. This includes threshold warning, trend increase warning, and trend speed increase warning. An alarm is triggered when a fault is detected. A three-stage noise reduction mechanism—sliding DC removal, current ripple notch filtering, and coherent environmental vibration filtering—significantly improves the signal-to-noise ratio of the vibration signal. Combined with envelope spectrum analysis and rotor dynamic tracking technology, bearing faults and rotor imbalances are accurately identified in strong noise environments, improving the fault identification rate.

[0100] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the motor fault detection method of this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0101] This application also provides a motor fault detection device; please refer to... Figure 6 The motor fault detection device includes: The acquisition module 10 is used to acquire the initial acceleration time-domain signal collected by the triaxial vibration sensor installed on the motor, the background vibration signal collected by the reference sensor installed on the motor base, and the three-phase current of the motor, wherein the triaxial vibration sensor is in rigid contact with the motor housing. Processing module 20 is used to perform sliding DC removal processing on the initial acceleration time-domain signal to obtain a reference acceleration time-domain signal; The processing module 20 is further configured to perform current notch filtering on the reference acceleration time-domain signal based on the three-phase current of the motor to obtain candidate acceleration time-domain signals; The filtering module 30 is used to perform environmental vibration coherent filtering on the candidate acceleration time-domain signal based on the background vibration signal to obtain the target acceleration time-domain signal; The filtering module 30 is also used to perform feature enhancement on the target acceleration time-domain signal to obtain an enhanced target acceleration time-domain signal; The fault detection module 40 is used to perform fault detection on the motor based on the enhanced target acceleration time-domain signal.

[0102] The motor fault detection device provided in this application, employing the motor fault detection method described in the above embodiments, can solve the technical problems of low detection efficiency and low accuracy in traditional motor fault detection methods. Compared with the prior art, the beneficial effects of the motor fault detection device provided in this application are the same as those of the motor fault detection method provided in the above embodiments, and other technical features of the motor fault detection device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0103] This application provides a motor fault detection device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the motor fault detection method in the above embodiment 1.

[0104] The following is for reference. Figure 7 The diagram illustrates a structural schematic suitable for implementing the motor fault detection device in the embodiments of this application. The motor fault detection device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 7 The motor fault detection device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0105] like Figure 7As shown, the motor fault detection device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in ROM (Read Only Memory) 1002 or a program loaded from storage device 1003 into RAM (Random Access Memory) 1004. RAM 1004 also stores various programs and data required for the operation of the motor fault detection device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touch screens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, LCDs (Liquid Crystal Displays), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the motor fault detection device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows motor fault detection devices with various systems, it should be understood that implementation or possession of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0106] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0107] The motor fault detection device provided in this application, employing the motor fault detection method described in the above embodiments, can solve the technical problems of low detection efficiency and low accuracy in traditional motor fault detection methods. Compared with the prior art, the beneficial effects of the motor fault detection device provided in this application are the same as those of the motor fault detection method provided in the above embodiments, and other technical features of this motor fault detection device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0108] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0109] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0110] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the motor fault detection method in the above embodiments.

[0111] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable Read Only Memory or Flash Memory), optical fibers, CD-ROM (CD-Read Only Memory), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0112] The aforementioned computer-readable storage medium may be included in the motor fault detection device; or it may exist independently and not assembled into the motor fault detection device.

[0113] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by the motor fault detection device, the motor fault detection device performs the following actions: acquires an initial acceleration time-domain signal collected by a triaxial vibration sensor mounted on the motor, a background vibration signal collected by a reference sensor mounted on the motor base, and the motor's three-phase current, wherein the triaxial vibration sensor is in rigid contact with the motor's housing; performs a sliding DC-free process on the initial acceleration time-domain signal to obtain a reference acceleration time-domain signal; performs current notch filtering on the reference acceleration time-domain signal based on the motor's three-phase current to obtain a candidate acceleration time-domain signal; performs environmental vibration coherent filtering on the candidate acceleration time-domain signal based on the background vibration signal to obtain a target acceleration time-domain signal; performs feature enhancement on the target acceleration time-domain signal to obtain an enhanced target acceleration time-domain signal; and performs fault detection on the motor based on the enhanced target acceleration time-domain signal.

[0114] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including LAN (Local Area Network) or WAN (Wide Area Network)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0115] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0116] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0117] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described motor fault detection method. This solves the technical problems of low detection efficiency and low accuracy in traditional motor fault detection methods. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the motor fault detection method provided in the above embodiments, and will not be repeated here.

[0118] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the motor fault detection method described above.

[0119] The computer program product provided in this application can solve the technical problems of low detection efficiency and low accuracy in traditional motor fault detection methods. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the motor fault detection method provided in the above embodiments, and will not be repeated here.

[0120] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method of detecting a fault in an electric machine, characterized by, The motor fault detection method comprises: obtaining an initial acceleration time domain signal collected by a three-axis vibration sensor installed on the motor, a background vibration signal collected by a reference sensor installed on the motor base, and three-phase currents of the motor, wherein the three-axis vibration sensor is in rigid contact with the shell of the motor; performing sliding DC removal processing on the initial acceleration time domain signal to obtain a reference acceleration time domain signal; performing current notch filtering on the reference acceleration time domain signal based on the three-phase currents of the motor to obtain a candidate acceleration time domain signal; performing environmental vibration coherence filtering on the candidate acceleration time domain signal based on the background vibration signal to obtain a target acceleration time domain signal; performing feature enhancement on the target acceleration time domain signal to obtain an enhanced target acceleration time domain signal; performing fault detection on the motor according to the enhanced target acceleration time domain signal.

2. The method of claim 1, wherein, The step of performing current notch filtering on the reference acceleration time domain signal based on the three-phase currents of the motor to obtain a candidate acceleration time domain signal comprises: obtaining a sampling frequency, a notch depth control parameter, and a time delay of the three-axis vibration sensor; extracting a current ripple characteristic frequency of the three-phase currents of the motor; performing current notch filtering on the reference acceleration time domain signal based on the sampling frequency, the notch depth control parameter, the time delay, and the current ripple characteristic frequency to obtain a candidate acceleration time domain signal.

3. The method of claim 2, wherein, The step of performing current notch filtering on the reference acceleration time domain signal based on the sampling frequency, the notch depth control parameter, the time delay, and the current ripple characteristic frequency to obtain a candidate acceleration time domain signal comprises: constructing a discrete-time transfer function of an adaptive notch filter based on the sampling frequency, the notch depth control parameter, the time delay, and the current ripple characteristic frequency; performing current notch filtering on the reference acceleration time domain signal according to the discrete-time transfer function to obtain a candidate acceleration time domain signal.

4. The method of claim 1, wherein, The step of performing sliding DC removal processing on the initial acceleration time domain signal to obtain a reference acceleration time domain signal comprises: obtaining a sampling frequency and a window point number of the three-axis vibration sensor; calculating a window duration according to the sampling frequency and the window point number; performing real-time sliding on the initial acceleration time domain signal based on the window duration to obtain a sliding acceleration time domain signal; calculating a signal average value of the sliding acceleration time domain signal within the window; performing DC removal processing on the initial acceleration time domain signal according to the signal average value to obtain a reference acceleration time domain signal.

5. The method of claim 1, wherein, The step of performing environmental vibration coherence filtering on the candidate acceleration time domain signal based on the background vibration signal to obtain a target acceleration time domain signal comprises: calculating a self-power spectrum of the background vibration signal and a self-power spectrum of the candidate acceleration time domain signal at a specific frequency; calculating a cross-power spectrum of the background vibration signal and the candidate acceleration time domain signal at the specific frequency; The candidate acceleration time domain signal is subjected to environmental vibration coherent filtering according to the cross power spectrum, the self power spectrum of the background vibration signal and the self power spectrum of the candidate acceleration time domain signal, to obtain a target acceleration time domain signal.

6. The method of claim 5, wherein, The step of subjecting the candidate acceleration time domain signal to environmental vibration coherent filtering according to the cross power spectrum, the self power spectrum of the background vibration signal and the self power spectrum of the candidate acceleration time domain signal, to obtain a target acceleration time domain signal, comprises: A linear coherence value between the background vibration signal and the candidate acceleration time domain signal is calculated according to the cross power spectrum, the self power spectrum of the background vibration signal and the self power spectrum of the candidate acceleration time domain signal; When the linear coherence value is greater than a preset coherence threshold value, the candidate acceleration time domain signal at the specific frequency is rejected, and the candidate acceleration time domain signal after rejection is taken as the target acceleration time domain signal.

7. The method of claim 1, wherein, The step of subjecting the motor to fault detection according to the enhanced target acceleration time domain signal comprises: An envelope spectrum is obtained according to the enhanced target acceleration time domain signal; A rated rotating speed of a motor rotor is obtained; A rotating speed critical value and a target rotating speed interval are determined according to the rated rotating speed; A target amplitude point is determined by searching in the target rotating speed interval in the envelope spectrum, with the rotating speed critical value as a starting point; An actual rotating speed of the motor rotor is determined according to the target amplitude point; The motor is subjected to fault detection based on the actual rotating speed and the envelope spectrum.

8. The method of claim 7, wherein, The step of subjecting the motor to fault detection based on the actual rotating speed and the envelope spectrum comprises: A bearing structure parameter is obtained; A theoretical fault characteristic frequency is calculated according to the actual rotating speed and the bearing structure parameter; It is detected whether there is a frequency peak value consistent with the theoretical fault characteristic frequency in the envelope spectrum; When it is detected that there is a frequency peak value consistent with the theoretical fault characteristic frequency in the envelope spectrum, it is determined that the bearing of the motor has a fault.

9. The method of any one of claims 1 to 8, wherein, The step of subjecting the target acceleration time domain signal to feature enhancement to obtain an enhanced target acceleration time domain signal comprises: The target acceleration time domain signal is subjected to signal transformation to obtain an analytic signal of the target acceleration time domain signal; An envelope signal is extracted according to the analytic signal; A power spectrum analysis is performed on the envelope signal to obtain an envelope spectrum; The enhanced target acceleration time domain signal is obtained according to the envelope spectrum.

10. An electric machine fault detection apparatus, characterized by, The device comprises: An obtaining module is configured to obtain an initial acceleration time domain signal collected by a three-axis vibration sensor installed on a motor, a background vibration signal collected by a reference sensor installed on a motor base and three-phase currents of the motor, wherein the three-axis vibration sensor is in rigid contact with a shell of the motor; A processing module is configured to perform sliding DC removal processing on the initial acceleration time domain signal to obtain a reference acceleration time domain signal; The processing module is further configured to perform current notch filtering on the reference acceleration time domain signal based on the three-phase currents of the motor to obtain a candidate acceleration time domain signal; The filter module is configured to perform environmental vibration coherent filtering on the candidate acceleration time-domain signal based on the background vibration signal to obtain a target acceleration time-domain signal. The filter module is further configured to perform feature enhancement on the target acceleration time-domain signal to obtain an enhanced target acceleration time-domain signal. The fault detection module is configured to perform fault detection on the motor according to the enhanced target acceleration time-domain signal.

Images