Mechanical fault diagnosis method and device, electronic equipment and computer readable medium

Abstract

The application relates to a mechanical fault diagnosis method and device, electronic equipment and a computer readable medium. The method comprises: performing generalized linear W transformation on a mechanical signal to generate a generalized linear W time-frequency spectrum of the mechanical signal; generating group time delays of the mechanical signal at each time-frequency position based on the generalized linear W time-frequency spectrum; generating a transient extraction generalized linear W time-frequency spectrum based on a transient extraction algorithm and the group time delays; and performing mechanical fault diagnosis based on the transient extraction generalized linear W time-frequency spectrum. The mechanical fault diagnosis method, device, electronic equipment and computer readable medium can more finely depict the non-stationary characteristics of a signal, extract the pulse characteristics of a mechanical fault signal, and thus efficiently identify bearing faults.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of mechanical fault diagnosis, and in particular, to a mechanical fault diagnosis method and device, an electronic device and a computer readable medium. BACKGROUND

[0002] Oilfield mechanical equipment has many types and large quantities, such as large compressors, pumps and other equipment. Due to large load variation, frequency conversion and harsh working conditions, the running state is relatively complex, and non-stationary processes exist in large quantities in mechanical equipment operation. Rolling bearings are widely used in large mechanical equipment in Shengli Oilfield and are one of the parts that are easily damaged.

[0003] When a rolling bearing fails, a periodic impact excitation will be generated, and the energy of the vibration signal will change. This change mainly manifests in the distribution of energy over time and frequency. The early vibration signal superimposes the response of this impact excitation, which makes the frequency spectrum indistinguishable, so it is necessary to find a detection method to separate the fault signal from the original signal. Therefore, it has certain practical significance to study the intelligent detection scheme of the early fault signal of mechanical equipment.

[0004] The operation of a rotating machine mainly relies on the rotational motion of the rotor system, and the rolling bearing is a supporting component of the rotor system, which becomes the core source of power for the rotating machine at a very high operating speed. The rolling bearing is mainly composed of a retainer, rolling elements, an inner ring and an outer ring, and the mutual cooperation of each part ensures the normal operation of the bearing, so any failure of any part may affect the state of the entire equipment.

[0005] According to data statistics, the rotating machine failures caused by bearings, rotors and gears account for 70% of the total number of failures, and among them, bearings alone account for more than 30%. Over the past few decades, great progress has been made in the fault diagnosis of rotating machines, especially rolling bearings, and many effective methods have been developed. Among them, the methods based on vibration information analysis have attracted widespread attention due to their non-destructive nature to the internal machinery and their high sensitivity to early faults. When the rolling elements roll over a defective surface or a defective element on the bearing surface, a pulse signal is generated, which is then amplified by the resonance of the bearing, and the signal can be captured by a vibration sensor installed on the bearing seat.

[0006] However, due to background noise and signal interference from other sources, bearing defects, especially early defects, are often undetectable, highlighting the need to develop effective signal processing techniques for accurate and reliable bearing fault diagnosis.

[0007] The above information disclosed in the background section is only used to enhance the understanding of the background of the present application, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0008] Therefore, the application provides a mechanical fault diagnosis method and device, electronic equipment and computer readable medium, which can more finely depict the non-stationary characteristics of the signal, extract the pulse characteristics of the mechanical fault signal, and efficiently identify bearing faults.

[0009] Other characteristics and advantages of the application will become apparent from the following detailed description, or will be learned by practice of the application.

[0010] According to an aspect of the application, a mechanical fault diagnosis method is provided, which includes: performing generalized linear W transform on a mechanical signal to generate a generalized linear W transform time-frequency spectrum of the mechanical signal; generating group time delay of the mechanical signal at each time-frequency position based on the generalized linear W transform time-frequency spectrum; generating a transient extraction generalized linear W transform time-frequency spectrum based on a transient extraction algorithm and the group time delay; and performing mechanical fault diagnosis based on the transient extraction generalized linear W transform time-frequency spectrum.

[0011] In an exemplary embodiment of the application, the generalized linear W transform is performed on the mechanical signal to generate the generalized linear W transform time-frequency spectrum, which includes: acquiring the mechanical signal of a bearing part based on a vibration sensor; and performing generalized linear W transform on the mechanical signal to generate the generalized linear W transform time-frequency spectrum.

[0012] In an exemplary embodiment of the application, the group time delay of the mechanical signal at each time-frequency position is generated based on the generalized linear W transform time-frequency spectrum, which includes: generating the group time delay of the mechanical signal at each time-frequency position based on phase information in the generalized linear W transform time-frequency spectrum.

[0013] In an exemplary embodiment of the application, the transient extraction generalized linear W transform time-frequency spectrum is generated based on the transient extraction algorithm and the group time delay, which includes: generating time-frequency coefficients based on the transient extraction algorithm and the group time delay; and generating the transient extraction generalized linear W transform time-frequency spectrum based on the time-frequency coefficients.

[0014] In an exemplary embodiment of the application, the time-frequency coefficients are generated based on the transient extraction algorithm and the group time delay, which includes: generating a transient extraction operator based on the transient extraction algorithm and the group time delay; and extracting the time-frequency coefficients at the delay curve in the generalized linear W transform time-frequency spectrum based on the transient extraction operator.

[0015] In an exemplary embodiment of the application, the transient extraction generalized linear W transform time-frequency spectrum is generated based on the time-frequency coefficients, which includes: performing modulo calculation on the time-frequency coefficients to generate the transient extraction generalized linear W transform time-frequency spectrum.

[0016] In an example embodiment of the present application, the target pulse is determined based on the transient extraction generalized linear W transform time-frequency spectrum, comprising: calculating a time-frequency envelope of the transient extraction generalized linear W transform time-frequency spectrum; taking a frequency corresponding to a maximum amplitude in the time-frequency envelope as a target frequency; and obtaining a target pulse corresponding to the target frequency.

[0017] In an example embodiment of the present application, the mechanical fault is determined based on a time-frequency feature of the target pulse, comprising: matching the time-frequency feature of the target pulse with a plurality of preset time-frequency features; and determining the fault of the pumping unit according to a matching result.

[0018] In an example embodiment of the present application, the method further comprises: generating the plurality of preset time-frequency features according to time-frequency data of pulses of a plurality of historical mechanical faults.

[0019] According to an aspect of the present application, a mechanical fault diagnosis apparatus is provided, comprising: a transform module configured to perform generalized linear W transform on a mechanical signal to generate a generalized linear W transform time-frequency spectrum of the mechanical signal; a time module configured to generate group time delays of the mechanical signal at each time-frequency position based on the generalized linear W transform time-frequency spectrum; an extraction module configured to generate a transient extraction generalized linear W transform time-frequency spectrum based on a transient extraction algorithm and the group time delays; and a diagnosis module configured to perform mechanical fault diagnosis based on the transient extraction generalized linear W transform time-frequency spectrum.

[0020] According to an aspect of the present application, an electronic device is provided, comprising: one or more processors; a storage device configured to store one or more programs; and when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described above.

[0021] According to an aspect of the present application, a computer readable medium having a computer program stored thereon is provided, and the program is executed by a processor to implement the method as described above.

[0022] According to the mechanical fault diagnosis method, apparatus, electronic device and computer readable medium of the present application, the generalized linear W transform is performed on the mechanical signal to generate a generalized linear W transform time-frequency spectrum of the mechanical signal; the group time delays of the mechanical signal at each time-frequency position are generated based on the generalized linear W transform time-frequency spectrum; the transient extraction generalized linear W transform time-frequency spectrum is generated based on the transient extraction algorithm and the group time delays; and the mechanical fault diagnosis is performed based on the transient extraction generalized linear W transform time-frequency spectrum. In this way, the non-stationary characteristics of the signal can be more finely depicted, the pulse features of the mechanical fault signal can be extracted, and the bearing fault can be efficiently identified.

[0023] It should be understood that the above general description and the following detailed description are only exemplary and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0024] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which:

[0025] Figure 1 is a system block diagram of a mechanical fault diagnosis method and device according to an exemplary embodiment.

[0026] Figure 2 is a flowchart of a mechanical fault diagnosis method according to an exemplary embodiment.

[0027] Figure 3 is an analog signal.

[0028] Figure 4 is a time-frequency spectrum obtained by performing STFT on the analog signal.

[0029] Figure 5 is a time-frequency spectrum obtained by performing S transform on the analog signal.

[0030] Figure 6 is a time-frequency spectrum obtained by performing generalized linear W transform on the analog signal.

[0031] Figure 7 is a time-frequency spectrum obtained by performing transient extraction generalized linear W transform on the analog signal.

[0032] Figure 8 is a waveform graph (a) and a spectrum graph (b) of a bearing outer ring fault signal.

[0033] Figure 9 is a time-frequency spectrum obtained by performing STFT on the bearing outer ring fault signal, and the right side is a partial enlarged view of the rectangular frame.

[0034] Figure 10 is a time-frequency spectrum obtained by performing transient extraction generalized linear W transform on the bearing outer ring fault signal, and the right side is a partial enlarged view of the rectangular frame.

[0035] Figure 11 is a block diagram of a mechanical fault diagnosis device according to another exemplary embodiment.

[0036] Figure 12 is a block diagram of an electronic device according to an exemplary embodiment.

[0037] Figure 13 is a block diagram of a computer readable medium according to an exemplary embodiment. DETAILED DESCRIPTION

[0038] Example embodiments now will be described more fully hereinafter with reference to the accompanying drawings. Example embodiments, however, can be implemented in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the example embodiments to those skilled in the art. Like reference numerals refer to like elements throughout the several views.

[0039] Moreover, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the application. One skilled in the relevant art will recognize, however, that the

[0040] The block diagrams in the drawings show only the functionality of the embodiments and do not imply any particular physical or architectural arrangement of the devices, systems, or methods. No inference should be drawn regarding the implementational aspects of the embodiments as shown and described herein. Further, having described a few embodiments, many variations and modifications will be suggested to those skilled in the art that are not specifically denoted herein. The description herein is to be considered in all respects only as illustrative and not restrictive.

[0041] The flow diagrams depicted herein are merely illustrative examples, and are not necessarily meant to imply a fixed order of operations, or that all of the operations or steps are to be performed. For example, certain operations / steps can be performed in a different order, or performed in parallel, or omitted, or combined with other operations / steps, depending on the circumstances. Also, the flow diagrams can not include all of the steps / stages that are necessary for the practice of the embodiments.

[0042] It should be understood that although the terms first, second, third, etc. can be used herein to describe various elements, components, regions, layers and / or sections, these elements, components, regions, layers and / or sections should not be limited by these terms. These terms are only used to distinguish one element, component, region, layer or section from another element, component, region, layer or section. Thus, a first element, component, region, layer or section discussed below could be termed a second element, component, region, layer or section without departing from the teachings of the present application. As used herein, the term "and / or" includes any and all combinations of one or more of the associated listed items.

[0043] Those skilled in the art will understand that the drawings described herein are merely illustrative and should not be construed as limiting the scope of the present application. The modules or processes in the drawings are not necessarily meant to be implemented in the described order, and should not be construed as limiting the scope of the present application.

[0044] The inventors of the present application have found that time-frequency analysis is a powerful tool for analyzing bearing faults and has received extensive attention over the past few decades. Classical linear methods, including short-time Fourier transform (STFT) and wavelet transform (CWT), can extend one-dimensional time series signals to two-dimensional time-frequency planes. From the time-frequency plane, we can observe the time-varying characteristics of the signal and perform signal decomposition. However, due to the Heisenberg uncertainty principle, the time-frequency representation generated by traditional methods is often ambiguous and cannot provide an accurate time-frequency description of time-varying signals. Recently, a new trend in the development of time-frequency analysis techniques is to use post-processing steps to represent the nonlinear characteristics of non-stationary signals, such as the reassignment method (RM), the synchrosqueezing transform (SST), and the synchrospectrogram transform (SET).

[0045] SST is a post-processing method that only reallocates time-frequency coefficients in the frequency (or scale) direction, and it can achieve perfect signal reconstruction. However, the first-order approximation used by SST when estimating the two-dimensional instantaneous frequency makes it insufficient to handle tonal signals. To address the above problems, various improved versions of SST have emerged, including second-order SST, matching SST, high-order SST, and multi-SST (MSST).

[0046] The inventors of the present application have found that the deficiency of SST lies in its lack of ability to remove noise interference, which is fundamentally due to the fact that the post-processing process of SST can redistribute noise into the final time-frequency representation. To address the above problems, the recently developed SET technique can retain the time-frequency coefficients that are most relevant to the original signal and discard the weakly relevant time-frequency coefficients to suppress the interference of noise.

[0047] For SST and SET, they both analyze the signal by assuming that the instantaneous frequency of the signal can be locally approximated as a time-invariant sequence within the analysis window. SST and SET are also extended to higher-order polynomial models for analyzing more complex signals. However, since the pulses generated by defects usually occur in a very short time and have a wideband frequency, the time-varying model cannot well approximate the pulse-like signal. This means that the time-varying assumption made by the above TFA techniques cannot well describe the signals generated by bearing faults. In order to more effectively analyze the pulse-like signal, He et al. proposed a time-reassignment SST (TSST), which only reassigns the time-frequency coefficients of the STFT along the time direction. Thereafter, Fourer and Auger extended it to the second-order TSST (TSST2). However, the above two methods failed to handle signals with nonlinear group delay (GD). In order to handle strong frequency-varying signals, He et al. proposed a time-reassignment MSST (TMSST), however, no matter how many times the TMSST is iterated, the time-frequency non-reassignment point always exists.

[0048] In view of the above problems in the prior art, in the mechanical fault diagnosis method provided in the present application, a mechanical fault diagnosis method based on high-precision transient state extraction generalized linear W transform can provide a high-precision GD estimation value, improve time-frequency energy aggregation, and realize high-precision diagnosis of a fault signal.

[0049] The content of the present application will be described in detail below with the help of specific embodiments.

[0050] Embodiment 1

[0051] Figure 1 It is a system block diagram of a mechanical fault diagnosis method, device, electronic equipment and computer readable medium according to an exemplary embodiment.

[0052] As shown in Figure 1 The system architecture 10 can include monitoring devices 101, 102, 103, a network 104 and a server 105. The network 104 is used to provide a communication link medium between the monitoring devices 101, 102, 103 and the server 105. The network 104 can include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0053] The monitoring devices 101, 102, 103 can be arranged near the pumping unit bearing and interact with the server 105 through the network 104 to receive or send messages, etc. Various monitoring type applications can be installed on the monitoring devices 101, 102, 103. The monitoring devices 101, 102, 103 can be various electronic devices with monitoring and wireless transmission functions, which are not limited by the present application.

[0054] The server 105 can be a server providing various services, such as a background management server monitoring signals transmitted by the monitoring devices 101, 102, and 103 for analysis. The background management server can perform analysis and the like on the received signals and feed back the processing results (e.g., failure analysis results) to an administrator.

[0055] The server 105 can perform, for example, a generalized linear W transform on a mechanical signal from the monitoring device 101, 102, and 103, generate a generalized linear W transform time-frequency spectrum of the mechanical signal; the server 105 can generate, for example, a group time delay of the mechanical signal at each time-frequency position based on the generalized linear W transform time-frequency spectrum; the server 105 can generate, for example, a transient extraction generalized linear W transform time-frequency spectrum based on a transient extraction algorithm and the group time delay; the server 105 can perform, for example, mechanical fault diagnosis based on the transient extraction generalized linear W transform time-frequency spectrum.

[0056] The server 105 can be a server of one entity, and can also be composed of multiple servers, for example. It should be noted that the mechanical fault diagnosis method provided in the embodiments of the present application can be executed by the server 105, and accordingly, the mechanical fault diagnosis apparatus can be arranged in the server 105. The application end for monitoring is generally located in the monitoring device 101, 102, and 103.

[0057] Embodiment 2

[0058] Figure 2 is a flowchart of a mechanical fault diagnosis method according to an exemplary embodiment. The mechanical fault diagnosis method 20 at least includes steps S202 to S208.

[0059] As shown in S202, a generalized linear W transform is performed on a mechanical signal to generate a generalized linear W transform time-frequency spectrum of the mechanical signal. For example, the mechanical signal of a bearing part can be acquired based on a vibration sensor; and the generalized linear W transform is performed on the mechanical signal to generate the generalized linear W transform time-frequency spectrum. Figure 2

[0060] In one specific embodiment, an original mechanical vibration signal x(t) can be acquired.

[0061] The original mechanical vibration signal x(t) is subjected to a generalized linear W transform to acquire a time-frequency spectrum of the generalized linear W transform GLWT(τ, f) of the original mechanical vibration signal x(t);

[0062] The time-frequency spectrum of the generalized linear W transform of the original mechanical vibration signal x(t) is:

[0063]

[0064] ​wherein t and f represent time and frequency respectively, τ is a time axis displacement parameter, i is an imaginary unit, and g(t, f; τ) represents a Gaussian window function; A and B are Gaussian window function adjustment factors;

[0065] The Gaussian window function is:

[0066]

[0067] wherein f0(τ)≠0 is a main frequency of the signal, Δf(τ)=f0(τ)-f; A and B are Gaussian window function adjustment factors; when the signal frequency f is equal to the main frequency f0(τ), the calculated time-frequency representation energy concentration is the highest.

[0068] In S204, group time delays of the mechanical signal at each time-frequency position are generated based on the generalized linear W transform time-frequency spectrum. For example, the group time delays of the mechanical signal at each time-frequency position can be generated based on phase information in the generalized linear W transform time-frequency spectrum.

[0069] The group time delay t x (τ, f) of the signal x(t) at each time-frequency position (τ, f) is estimated according to the phase information of the time-frequency spectrum.

[0070] The group time delay t x (τ, f) of the signal x(t) at each time-frequency position (τ, f) is estimated according to the phase information of the generalized linear W transform time-frequency spectrum.

[0071]

[0072] wherein is a partial derivative symbol, represents a partial derivative of GLWT(τ, f) with respect to frequency f; GLWT g′ (τ, f) is a GLWT result calculated using .

[0073] In S206, a transient extraction generalized linear W transform time-frequency spectrum is generated based on a transient extraction algorithm and the group time delays. For example, time-frequency coefficients can be generated based on the transient extraction algorithm and the group time delays; and the transient extraction generalized linear W transform time-frequency spectrum can be generated based on the time-frequency coefficients.

[0074] In one embodiment, the time-frequency coefficients are generated based on the transient extraction algorithm and the group time delays, including: generating a transient extraction operator based on the transient extraction algorithm and the group time delays; and extracting time-frequency coefficients at a delay curve in the generalized linear W transform time-frequency spectrum based on the transient extraction operator.

[0075] In one embodiment, the transient extraction generalized linear W transform time-frequency spectrum is generated based on the time-frequency coefficients, including: performing a modulo calculation on the time-frequency coefficients to generate the transient extraction generalized linear W transform time-frequency spectrum.

[0076] According to the principle of transient extraction, a transient extraction operator (TEO) centered on the signal group delay curve is constructed in the time-frequency domain: δ(τ-t x (τ,f)) is used to extract the time-frequency coefficients at the group delay curve of the original time-frequency spectrum, and a transient extraction generalized linear W transform value Te(τ,f) is obtained;

[0077] The time-frequency spectrum of the mechanical vibration signal x(t) after the transient extraction generalized linear W transform is:

[0078] Te(τ,f)=S GLWT δ(τ-t x (τ,f)) (4)

[0079] Where δ(t) is the unit impulse function, and δ(τ-t x (τ,f)) represents the transient extraction operator in the GLWT time-frequency domain.

[0080] In S208, mechanical fault diagnosis is performed based on the transient extraction generalized linear W transform time-frequency spectrum. For example, a target pulse can be determined based on the transient extraction generalized linear W transform time-frequency spectrum; and a mechanical fault can be determined based on the time-frequency characteristics of the target pulse.

[0081] In one embodiment, the target pulse is determined based on the transient extraction generalized linear W transform time-frequency spectrum, including: calculating the time-frequency envelope of the transient extraction generalized linear W transform time-frequency spectrum; taking the frequency corresponding to the maximum amplitude in the time-frequency envelope as the target frequency; and obtaining the target pulse corresponding to the target frequency.

[0082] The modulus of Te(τ,f) is taken to obtain the time-frequency spectrum of the transient extraction generalized linear W transform.

[0083] In one embodiment, the mechanical fault is determined based on the time-frequency characteristics of the target pulse, including: matching the time-frequency characteristics of the target pulse with a plurality of preset time-frequency characteristics; and determining the fault of the pumping unit according to the matching result. The pulse characteristics of the frequency corresponding to the strongest amplitude in the obtained time-frequency spectrum are extracted, and the time interval of the pulse characteristics is compared with the matching degree between the known various fault signals to identify the bearing fault of the pumping unit.

[0084] According to the mechanical fault diagnosis method, the generalized linear W transform time-frequency spectrum of the mechanical signal is generated by performing generalized linear W transform on the mechanical signal; the group time delay of the mechanical signal at each time-frequency position is generated based on the generalized linear W transform time-frequency spectrum; and the transient extraction generalized linear W transform time-frequency spectrum is generated based on the transient extraction algorithm and the group time delay.

[0085] The working principle of the mechanical fault diagnosis method is as follows: collecting a mechanical vibration signal x(t) to be analyzed; calculating the time-frequency spectrum of the generalized linear W transform of the original mechanical vibration signal x(t) by using the generalized linear W transform; estimating the group time delay of the signal x(t) at each time-frequency position (tau, f) according to the phase information of the time-frequency spectrum; constructing a transient extraction operator centered on the group time delay curve of the original time-frequency spectrum on the time-frequency domain to extract the time-frequency coefficients at the group time delay curve of the original time-frequency spectrum according to the transient extraction principle, and obtaining the transient extraction generalized linear W transform value; and taking the modulus of the obtained time-frequency coefficients to obtain the time-frequency spectrum of the transient extraction generalized linear W transform. The mechanical fault diagnosis method can significantly extract the pulse characteristics of the frequency corresponding to the strong amplitude of the mechanical fault signal, and identify the bearing fault by comparing the time interval of the pulse characteristics.

[0086] Compared with the prior art, the mechanical fault diagnosis method has the following advantages:

[0087] (1) The basic theory of the transient extraction generalized linear W transform is given, and the expression of the transient extraction generalized linear W transform is derived;

[0088] (2) The transient extraction generalized linear W transform is a new time-frequency analysis method, which combines the advantages of the generalized linear W transform and the transient extraction transform, and has high time-frequency focusing;

[0089] (3) The transient extraction generalized linear W transform is different from the W transform. The transient extraction generalized linear W transform can flexibly adjust the change trend of the Gaussian window function of the transient extraction generalized linear W transform by adjusting two different parameters according to actual needs, so as to more flexibly adapt to the analysis and processing of specific signals, and the transient extraction generalized linear W transform has better anti-aliasing performance;

[0090] (4) The transient extraction generalized linear W transform is different from the transient extraction transform. The transient extraction transform is based on the extraction operation of the short-time Fourier transform, while the transient extraction generalized linear W transform is based on the extraction operation of the generalized linear W transform. The transient extraction generalized linear W transform can more finely depict the non-stationary characteristics of the signal, extract the pulse characteristics of the mechanical fault signal, and thus efficiently identify the bearing fault.

[0091] It should be clearly understood that the application describes how to form and use particular examples, but the principles of the application are not limited to any details of these examples. Rather, these principles can be applied to many other embodiments based on the teachings of the disclosure of the application.

[0092] Example 3

[0093] Take an analog fault signal with pulse component as an example, the waveform of the signal is shown in Figure 3 . Figures 4-7 are the STFT time-frequency spectrum, S-transform time-frequency spectrum, generalized linear W-transform time-frequency spectrum and transient-extraction generalized linear W-transform time-frequency spectrum of the signal respectively. In the figure, the horizontal axis Time / s represents time, and the vertical axis Frequency / Hz represents frequency. The time-frequency resolution of the analog fault signal calculated by the STFT is low. The energy at the main frequency of the time-frequency spectrum based on the S-transform exists a phenomenon of shifting to a higher frequency, and the S-transform shows a low resolution at the low frequency. The generalized linear W-transform significantly improves the time resolution at the low frequency end, and makes the energy of the time-frequency spectrum focus on the main frequency of the signal. The transient-extraction generalized linear W-transform can accurately extract the time-frequency characteristics of the pulse signal, more finely depict the non-stationary characteristics of the signal, and make the energy of the time-frequency spectrum focus on the main frequency of the signal. This characteristic is of great help to accurately identify the occurrence of the fault position of the rolling body impacting the bearing system.

[0094] As shown in Figures 8-10 . In the experiment, the vibration signal of the fault bearing was collected by using an acceleration sensor, and the sensor was placed at the driving end of the motor shell. At this time, the sampling frequency was 12 kHz, and the working speed of the motor shaft was 1797 rpm. According to the bearing parameters and the speed, the characteristic frequency of the outer ring fault was 107.4 Hz. The collected outer ring fault vibration signal and its spectrum diagram are plotted in Figure 8 . It can be seen that during the rotation of the fault bearing, a periodic pulse signal is generated. Due to serious noise interference, it is difficult to accurately locate the position of these transient signals in the time domain signal. However, the spectral analysis shows that the time-frequency information of the signal is mainly concentrated in the frequency range of 2.5-4.0 kHz. Figure 9 and Figure 10 are the time-frequency representation results calculated by using the STFT and the transient-extraction generalized linear W-transform respectively, and the right side shows the local magnified diagram corresponding to 40-47 ms. With the help of time-frequency analysis technology, the time-frequency characteristics of repeated transients can be expanded to the time-frequency plane to provide more important information than the time domain analysis alone. It can be observed that the STFT provides a rough time-frequency description of the transient signal, making it difficult to accurately locate the time-frequency information of the signal. The transient-extraction generalized linear W-transform can accurately capture the pulse characteristics, which is helpful for accurately diagnosing the characteristic frequency of the bearing fault. From Figure 9From this, we can know that the time interval between two ring transient is 9.33ms. Therefore, according to the inverse of this value, the fault feature frequency can be calculated as 107.2Hz. This value is very close to the theoretical value, indicating that this method can effectively diagnose bearing faults.

[0095] Those skilled in the art can understand that all or part of the steps of the above-mentioned embodiments are implemented as a computer program executed by a CPU. When the computer program is executed by the CPU, the above-mentioned functions defined by the above-mentioned method provided by the present application are executed. The program can be stored in a computer readable storage medium, which can be a read-only memory, a disk or an optical disk, etc.

[0096] In addition, it should be noted that the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present application, and are not for limiting purposes. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be executed synchronously or asynchronously, for example, in multiple modules.

[0097] Embodiment 4

[0098] The following is a device embodiment of the present application, which can be used to execute the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0099] Figure 11 is a block diagram of a mechanical fault diagnosis device according to an exemplary embodiment. As shown in Figure 11 , the mechanical fault diagnosis device 110 includes a transformation module 1102, a time module 1104, an extraction module 1106, and a diagnosis module 1108.

[0100] The transformation module 1102 is configured to perform generalized linear W transformation on a mechanical signal to generate a generalized linear W time-frequency spectrum of the mechanical signal. The transformation module 1102 is further configured to acquire the mechanical signal of a bearing part based on a vibration sensor, and perform generalized linear W transformation on the mechanical signal to generate the generalized linear W time-frequency spectrum.

[0101] The time module 1104 is configured to generate a group time delay of the mechanical signal at each time-frequency position based on the generalized linear W time-frequency spectrum. The time module 1104 is further configured to generate a group time delay of the mechanical signal at each time-frequency position based on phase information in the generalized linear W time-frequency spectrum.

[0102] The extraction module 1106 is configured to generate a transient extraction generalized linear W time-frequency spectrum based on a transient extraction algorithm and a group time delay. The extraction module 1106 is further configured to generate a time-frequency coefficient based on the transient extraction algorithm and the group time delay, and generate the transient extraction generalized linear W time-frequency spectrum based on the time-frequency coefficient.

[0103] The diagnosis module 1108 is configured to perform mechanical fault diagnosis based on the transient extraction generalized linear W transform time-frequency spectrum. The diagnosis module 1108 is further configured to determine a target pulse based on the transient extraction generalized linear W transform time-frequency spectrum; and determine mechanical fault based on time-frequency features of the target pulse.

[0104] According to the mechanical fault diagnosis apparatus, the generalized linear W transform time-frequency spectrum of the mechanical signal is generated by performing generalized linear W transform on the mechanical signal; the group time delay of the mechanical signal at each time-frequency position is generated based on the generalized linear W transform time-frequency spectrum; the transient extraction generalized linear W transform time-frequency spectrum is generated based on the transient extraction algorithm and the group time delay; and the mechanical fault diagnosis is performed based on the transient extraction generalized linear W transform time-frequency spectrum. In this way, the non-stationary characteristics of the signal can be more finely described, the pulse features of the mechanical fault signal can be extracted, and the bearing fault can be efficiently identified.

[0105] Embodiment 5

[0106] Figure 12 is a block diagram of an electronic device according to an exemplary embodiment.

[0107] The electronic device 1200 according to this embodiment of the present application will be described below with reference to Figure 12 FIG. 1. Figure 12 The electronic device 1200 shown is merely an example and should not limit the function and scope of use of the embodiments of the present application.

[0108] As shown in Figure 12 FIG. 1, the electronic device 1200 is in the form of a general computing device. The components of the electronic device 1200 can include, but are not limited to, at least one processing unit 1210, at least one storage unit 1220, a bus 1230 connecting different system components (including the storage unit 1220 and the processing unit 1210), a display unit 1240, and the like.

[0109] The storage unit stores program codes which can be executed by the processing unit 1210, so that the processing unit 1210 performs the steps described in the present specification according to various exemplary embodiments of the present application. For example, the processing unit 1210 can perform the steps as shown in Figure 2 FIG. 1.

[0110] The storage unit 1220 can include a readable medium in the form of a volatile storage unit, such as a random access memory (RAM) 12201 and / or a cache memory 12202, and can further include a read-only memory (ROM) 12203.

[0111] The storage unit 1220 can also include a program / utility 12204 having a set of programs / modules 12205, each of which performs one or more of the operations described herein, and / or some combination thereof. The programs 12205 can include, for example, boot-up programs, application programs, other program modules, and program data, each of which can include an operating system. Each of the operating system, one or more application programs, other program modules, and program data can include, in whole or in part, by way of example, implementations of the network environment.

[0112] The bus 1230 can represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration bus, a processor or local bus using any of a variety of bus architectures, and so forth.

[0113] The electronic device 1200 can also communicate with one or more external devices 1200' such as a keyboard, a pointing device, a Bluetooth device, etc. so as to facilitate user interaction with the electronic device 1200. Additionally, the electronic device 1200 can communicate with one or more other computing devices, such as a router, a modem, etc. so as to facilitate communication with one or more other computing devices. Such communication can be facilitated via an input / output (I / O) interface 1250. Further, the electronic device 1200 can communicate with one or more networks, such as a local area network (LAN), a wide area network (WAN), and / or the public network, such as the Internet, via a network adapter 1260. The network adapter 1260 can communicate with the other components of the electronic device 1200 via the bus 1230. It should be appreciated that although the network adapter 1260 is illustrated as a single component, the network adapter 1260 can comprise two or more components that work together to facilitate communications with one or more other computing devices.

[0114] From the foregoing description, it will be apparent to a person skilled in the art that the example embodiments described herein can be implemented in software and / or can be implemented as software in combination with the necessary hardware. Therefore, as Figure 13 shown, the technical solutions according to the embodiments of the present application can be embodied in the form of a software product. The software product can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.) or a network, and includes a plurality of instructions to make a computing device (which can be a personal computer, a server, or a network device, etc.) execute the above-mentioned method according to the embodiments of the present application.

[0115] The software product can employ any combination of one or more computer-readable media. The computer-readable media can be a computer-readable storage medium or a computer-readable signal medium. The computer-readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include the following: an electrical connection having one or more wires, a portable disc, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0116] The computer-readable storage medium can include data signals on a carrier wave modulated or otherwise propagated on a propagation medium. The propagation medium can be any medium that can carry and propagate computer program code, including, but not limited to, electromagnetic, optical, or any suitable combination of the foregoing. The computer-readable storage medium can also be any medium that can be used to store the desired program code, including, but not limited to, a floppy disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0117] The program code can be executed by one or more programmable processors, which can be implemented using one or more microprocessors, microcontrollers, digital signal processors, application specific integrated circuits, field programmable gate arrays, programmable logic devices, or any other suitable programmable device. The program code can be stored on any suitable computer readable medium, including one or more memories (e.g., removable, on-board, or embedded memories) associated with the computing device. In some embodiments, the program code can be downloaded to the computing device from the Internet or another network. In other embodiments, the program code can be downloaded to the computing device from a removable computer readable storage medium. In some embodiments, the program code can be downloaded to the computing device from a combination of the Internet and the removable computer readable storage medium. In some embodiments, the program code can be downloaded to the computing device from a combination of the Internet and the removable computer readable storage medium.

[0118] The computer readable medium carries one or more programs, when the one or more programs are executed by the device, the computer readable medium enables the following functions: performing a generalized linear W transform on a mechanical signal to generate a generalized linear W transform time-frequency spectrum of the mechanical signal; generating a group time delay of the mechanical signal at each time-frequency position based on the generalized linear W transform time-frequency spectrum; generating a transient extraction generalized linear W transform time-frequency spectrum based on a transient extraction algorithm and the group time delay; and performing mechanical fault diagnosis based on the transient extraction generalized linear W transform time-frequency spectrum.

[0119] Those skilled in the art can understand that the above modules can be distributed in the device according to the description of the embodiments, and can also be changed in one or more devices different from the embodiments. The modules of the above embodiments can be combined into one module, or further split into multiple sub-modules.

[0120] Through the above description of the embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solutions according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a U disk, a mobile hard disk, etc.) or a network, and includes a plurality of instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the embodiments of the present application.

[0121] The example embodiments of the present application are specifically shown and described above. It should be understood that the present application is not limited to the detailed structure, arrangement or implementation method described herein; on the contrary, the present application is intended to cover various modifications and equivalent arrangements included in the spirit and scope of the appended claims.

Claims

1. A mechanical failure diagnosis method characterized by, The method comprises the following steps: performing generalized linear W transform on the mechanical signal to generate a generalized linear W transform time-frequency spectrum of the mechanical signal; generating group time delay of the mechanical signal at each time-frequency position based on the generalized linear W transform time-frequency spectrum; generating a transient extraction generalized linear W transform time-frequency spectrum based on the transient extraction algorithm and the group time delay; performing mechanical fault diagnosis based on the transient extraction generalized linear W transform time-frequency spectrum; The method comprises the following steps: obtaining an original mechanical vibration signal x(t), performing generalized linear W transform on the original mechanical vibration signal x(t) to obtain a time-frequency spectrum of the generalized linear W transform GLWT(τ, f) of the original mechanical vibration signal x(t); wherein the time-frequency spectrum of the generalized linear W transform of the original mechanical vibration signal x(t) is: wherein t and f represent time and frequency respectively, τ is a time axis displacement parameter, i is an imaginary unit, and a Gaussian window function is represented as g(t, f; τ); A and B are Gaussian window function adjustment factors; wherein f0(τ)≠0 is a main frequency of the signal, Δf(τ)=f0(τ)-f; A and B are Gaussian window function adjustment factors; when the signal frequency f is equal to the main frequency f0(τ), the time-frequency representation calculated at this time has the highest energy concentration; generating group time delay of the mechanical signal at each time-frequency position based on the generalized linear W transform time-frequency spectrum, comprising: The group delay t of the signal x(t) at each time-frequency position (τ, f) is estimated from the phase information of the time-frequency spectrum of the generalized linear W transform GLWT(τ, f) x (τ, f): wherein is the partial derivative symbol, f GLWT(τ, f) denotes the partial derivative of GLWT(τ, f) with respect to the frequency f; GLWT g′ (τ, f) is the GLWT result computed using the partial derivative of GLWT(τ, f) with respect to the frequency f; GLWT generating a transient extraction generalized linear W transform time-frequency spectrum based on the transient extraction algorithm and the group time delay, comprising: According to the principle of transient extraction, a transient extraction operator δ(τ-t x (τ, f)) is used to extract the time-frequency coefficients at the group time-delay curve of the original time-frequency spectrum, and a transient extraction generalized linear W transform value Te(τ, f) is obtained. wherein the time-frequency spectrum of the transient extraction generalized linear W transform of the mechanical vibration signal x(t) is: where δ(t) is the unit impulse function, δ(τ-t x (τ, f)) represents the transient extraction operator in the GLWT time-frequency domain; taking the modulus of Te(τ, f) to obtain the time-frequency spectrum of the transient extraction generalized linear W transform; performing mechanical fault diagnosis based on the transient extraction generalized linear W transform time-frequency spectrum, comprising: determining a target pulse based on the transient extraction generalized linear W transform time-frequency spectrum; determining a mechanical fault based on the time-frequency characteristics of the target pulse; determining a target pulse based on the transient extraction generalized linear W transform time-frequency spectrum, comprising: calculating a time-frequency envelope of the transient extraction generalized linear W transform time-frequency spectrum; taking the frequency corresponding to the maximum amplitude in the time-frequency envelope as a target frequency; obtaining a target pulse corresponding to the target frequency; determining a mechanical fault based on the time-frequency characteristics of the target pulse, comprising: matching the time-frequency characteristics of the target pulse with a plurality of preset time-frequency characteristics, wherein the plurality of preset time-frequency characteristics are generated according to the time-frequency data of pulses of a plurality of historical mechanical faults; determining a mechanical fault according to the matching result.

2. A mechanical failure diagnosing apparatus characterized by comprising: The method comprises the following steps: a transform module, configured to perform generalized linear W transform on a mechanical signal to generate a generalized linear W transform time-frequency spectrum of the mechanical signal; The method comprises the following steps: obtaining an original mechanical vibration signal x(t), performing generalized linear W transform on the original mechanical vibration signal x(t) to obtain a time-frequency spectrum of the generalized linear W transform GLWT(τ, f) of the original mechanical vibration signal x(t); wherein the time-frequency spectrum of the generalized linear W transform of the original mechanical vibration signal x(t) is: In the formula, t and f respectively represent time and frequency, τ is a time axis displacement parameter, i is an imaginary unit, a Gaussian window function is represented as g(t, f; τ), and A and B are Gaussian window function adjustment factors. In the formula, f0(τ)≠0 is a main frequency of the signal, Δf(τ)=f0(τ)-f, A and B are Gaussian window function adjustment factors, and when the signal frequency f is equal to the main frequency f0(τ), the calculated time-frequency representation has the highest energy concentration; a time module configured to generate group time delay of the mechanical signal at each time-frequency position based on the generalized linear W transform time-frequency spectrum; generating group time delay of the mechanical signal at each time-frequency position based on the generalized linear W transform time-frequency spectrum, comprises: The group delay t of the signal x(t) at each time-frequency position (τ, f) is estimated from the phase information of the time-frequency spectrum of the generalized linear W transform GLWT(τ, f) x (τ, f): wherein is the partial derivative symbol, f GLWT(τ, f) denotes the partial derivative of GLWT(τ, f) with respect to the frequency f; GLWT g′ (τ, f) is the GLWT result computed using the partial derivative of GLWT(τ, f) with respect to the frequency f; GLWT a extracting module configured to generate a transient extraction generalized linear W transform time-frequency spectrum based on a transient extraction algorithm and the group time delay; generating a transient extraction generalized linear W transform time-frequency spectrum based on a transient extraction algorithm and the group time delay, comprises: According to the principle of transient extraction, a transient extraction operator δ(τ-t x (τ, f)) is constructed in time-frequency domain with the signal group delay curve as the center to extract the time-frequency coefficients at the group delay curve of the original time-frequency spectrum, and a generalized linear W transform value Te(τ, f) of the transient extraction is obtained. wherein the time-frequency spectrum of the mechanical vibration signal x(t) after the transient extraction generalized linear W transform is: where δ(t) is the unit impulse function, δ(τ - t x (τ, f) represents the transient extraction operator in the GLWT time-frequency domain; taking the modulus of Te(τ, f) to obtain the transient extraction generalized linear W transform time-frequency spectrum; a diagnosing module configured to perform mechanical fault diagnosis based on the transient extraction generalized linear W transform time-frequency spectrum; performing mechanical fault diagnosis based on the transient extraction generalized linear W transform time-frequency spectrum, comprises: determining a target pulse based on the transient extraction generalized linear W transform time-frequency spectrum; determining a mechanical fault based on a time-frequency feature of the target pulse; determining a target pulse based on the transient extraction generalized linear W transform time-frequency spectrum, comprises: calculating a time-frequency envelope of the transient extraction generalized linear W transform time-frequency spectrum; taking a frequency corresponding to a maximum amplitude in the time-frequency envelope as a target frequency; obtaining a target pulse corresponding to the target frequency; determining a mechanical fault based on a time-frequency feature of the target pulse, comprises: matching the time-frequency feature of the target pulse with a plurality of preset time-frequency features, wherein the plurality of preset time-frequency features are generated according to time-frequency data of pulses of a plurality of historical mechanical faults; determining the mechanical fault according to a matching result.

3. An electronic device, comprising: comprises: one or more processors; a storage device configured to store one or more programs; when the one or more programs are executed by the one or more processors, the one or more processors implement the method of claim 1.

4. A computer readable medium having stored thereon a computer program, characterized in that, the program is executed by the processor to implement the method of claim 1. the program is executed by the processor to implement the method of claim 1.

Images