Rotary machinery vibration signal normalization method and device and storage medium

Abstract

The invention discloses a rotary machinery vibration signal normalization method and device and a storage medium, and the method comprises the steps: carrying out the Hilbert transformation of a collected non-stationary vibration signal, and obtaining an envelope signal; based on the slope, dividing the envelope signal into a plurality of sub-signals with linearly changed speeds; respectively carrying out frequency domain normalization on each sub-signal; solving an amplitude normalization function; and calculating the root-mean-square characteristic of the normalized signal, and diagnosing the running state of the rotating machine under the non-stationary working condition. According to the method, the non-stationary vibration signals can be normalized in the frequency domain and the time domain, interference of rotating speed fluctuation on vibration characteristics is reduced, and fault diagnosis of the rotating machine under the non-stationary working condition is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of rotating machinery vibration signal processing and diagnosis under non-stationary working conditions, and in particular to a rotating machinery vibration signal normalization method, device and storage medium. Background Art

[0002] Rotating machinery plays an indispensable role in various fields, including automotive, marine, and aerospace. Due to harsh operating environments and changing conditions, localized damage to key components in rotating machinery is inevitable. Therefore, health monitoring and fault detection are crucial to ensuring safe machine operation and reducing maintenance costs. In industrial applications, rotating machinery often operates under non-stationary conditions. Speed fluctuations induce amplitude modulation in the time domain of the vibration signal and frequency smearing in the spectrum. However, localized component failures can also cause similar modulation effects. Speed fluctuations can interfere with the fault diagnosis process and cause false fault alarms. Therefore, it is necessary to eliminate the interference of speed fluctuations and improve the diagnostic accuracy of rotating machinery under non-stationary conditions.

[0003] In recent years, some researchers have tried to eliminate the amplitude modulation effect caused by speed fluctuations, but several problems still exist. In engineering practice, the speed change process of rotating machinery is continuous and nonlinear, so the collected vibration signal exhibits non-stationary characteristics. If the existing time-frequency transformation method is directly applied to the original signal, the corresponding time-frequency representation around the speed trend change point will become blurred. Although the existing order tracking algorithm can suppress the frequency tailing in the spectrum, when the speed fluctuates violently, angle resampling will cause envelope deformation. Speed fluctuations and local damage of rotating machinery will produce an amplitude modulation effect on the vibration signal. The existing normalization method cannot accurately eliminate the amplitude modulation caused by speed while retaining the fault feature information. Summary of the Invention

[0004] In order to solve the current technical problems, the main purpose of the present invention is to provide a method, device and storage medium for normalizing the vibration signal of a rotating machinery to eliminate the modulation effect caused by speed fluctuations, thereby retaining the fault characteristic information in the vibration signal.

[0005] In order to overcome the problems existing in the prior art, the technical solution adopted by the present invention is: a method for normalizing the vibration signal of a rotating machine, comprising the following steps: Step S1, performing Hilbert transform on the collected non-stationary vibration signal to obtain an envelope signal; Step S2: based on the slope, dividing the envelope signal into a plurality of sub-signals with linearly varying speeds; Step S3: performing frequency domain normalization on each sub-signal; Step S4, solving the amplitude normalization function; Step S5: Calculate the root mean square characteristics of the normalized signal to diagnose the operating state of the rotating machinery under non-stationary working conditions. In step S1, the original non-stationary vibration signal is subjected to Hilbert transform to obtain the envelope signal, and the calculation formula is as follows: ; Where: represents the original vibration signal collected; τ is the integration variable, which is used to represent the value point of the signal x(τ) during the integration process; Represents the envelope signal.

[0006] In step S2, the envelope signal is segmented. The length of each segment is half the sampling frequency. The slope of each segment is calculated. Segments with similar slopes form a sub-signal. The signal under nonlinear speed change conditions is divided into multiple sub-signals with linear speed changes. The slope calculation formula is as follows: ; Where: Indicates that the length of each small segment of the signal is half the sampling frequency, Indicates that the original envelope signal is divided into part, Indicates the The mean value of the time axis of the segment envelope signal, Indicates the The mean value of the segment envelope signal amplitude, Indicates that the length of each segment is W The time variable of the envelope signal, Indicates the amplitude of each small segment of the envelope signal, Indicates slope calculation.

[0007] In step 3, the sub-signals are normalized in the frequency domain using synchronized time warping.

[0008] The synchronous time warping based on the phase function is used to adjust the scale of the time axis of the non-stationary vibration signal to achieve non-uniform resampling. The formula is as follows: ; Where, Indicates the k The inverse of the phase function of the sub-signal, Indicates the k The result after non-uniform resampling of the sub-signals is: Indicates the k The amplitude parameters of the sub-signals, represents the correction coefficient, A time variable representing the time-varying signal, Indicates the k The phase function of the sub-signal; Then, by finding a unified fundamental frequency, the characteristic frequencies under different speed conditions are aligned. The formula is as follows: ; Where, represents the fundamental frequency, Indicates the k The characteristic ridge line after the sub-signal is resampled, Indicates taking the maximum value, Indicates taking the minimum value.

[0009] In step 4, the amplitude normalization function is solved by power fitting to achieve time domain normalization.

[0010] The amplitude normalization function is solved by power fitting, and the formula for time domain normalization is as follows: ; ; Where: Indicates the speed signal, represents the normalization function, and represents the linear fitting parameter, represents the power fitting parameter, Indicates the k The result after resampling the sub-signals is represents the normalized signal.

[0011] In step 5, the root mean square characteristic curve of the normalized signal is calculated to determine whether the current operating state of the rotating machinery is faulty. The root mean square calculation formula is as follows: ; Where: n Indicates the signal length, represents the normalized signal; For the root mean square characteristic curve of the normalized signal, it is determined whether a fault occurs according to the set alarm threshold.

[0012] A device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a rotating machinery vibration signal normalization method is implemented.

[0013] A storage medium stores a computer program, and when the computer program is executed by a processor, the method for normalizing a rotating machinery vibration signal is implemented.

[0014] The present invention has the following beneficial effects: This invention utilizes a slope-based segmentation method to divide a nonlinear velocity variation process into multiple linear velocity sub-signals. By adjusting the time axis scale and aligning the characteristic frequencies of the different sub-signals, synchronized time warping is achieved for frequency normalization. Approximating the amplitude modulation effect as a normalized function eliminates the amplitude modulation effect while preserving fault signature information. The root mean square (RMS) characteristics of the normalized signals can be used to accurately diagnose the current operating status of rotating machinery. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 It is a flow chart of the present invention.

[0017] Figure 2 It is a time domain diagram of the original vibration signal of the planetary gearbox under non-stationary working conditions in the present invention.

[0018] Figure 3 It is an envelope signal diagram of the original vibration signal in the present invention.

[0019] Figure 4 This is a diagram showing the result of segmenting the envelope signal based on the slope in the present invention.

[0020] Figure 5 It is a time-frequency representation diagram of the signal after synchronous time distortion in the present invention.

[0021] Figure 6 This is a diagram of the processing result of the normalization function in the time domain in the present invention.

[0022] Figure 7 It is a root mean square characteristic curve diagram of the normalized signal in the present invention. DETAILED DESCRIPTION

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0024] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0025] Secondly, the "present embodiment" or "embodiment" referred to herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention.

[0026] Example 1: In this embodiment, a method for normalizing a vibration signal of a rotating machine is provided. Figure 1 The present invention is a flowchart of a method for normalizing a rotating machinery vibration signal.

[0027] like Figure 1 As shown, a rotating machinery vibration signal normalization method involved in this embodiment includes the following steps: Step S1: Perform Hilbert transform on the original non-stationary vibration signal to obtain an envelope signal.

[0028] Figure 2 It is a time domain diagram of the original vibration signal of the planetary gearbox under non-stationary working conditions in the present invention.

[0029] Figure 3 It is an envelope signal diagram of the original vibration signal in the present invention.

[0030] In this embodiment, an acceleration sensor is placed on the planetary gearbox test bench to collect the vibration signal of the planetary gearbox under non-stationary working conditions in real time. The sampling frequency is 20000 Hz. After the Hilbert transform of the original non-stationary signal, the envelope signal is expressed as follows: ; Where, Represents the collected original vibration signal, τ Is the integral variable used to represent the signal during the integration process x ( τ ) value point, Represents the envelope signal. Figure 2 As shown in Figure 2, the vibration signal collected under non-stationary working conditions has obvious amplitude modulation phenomenon, and the amplitude of the signal changes with the trend of speed change. Figure 3 As shown in Figure 2, the pulse features in the envelope signal are more obvious than the original vibration signal.

[0031] Step S2: Based on the slope, the envelope signal is divided into a plurality of sub-signals with linearly varying speeds.

[0032] Figure 4 This is a diagram showing the result of segmenting the envelope signal based on the slope in the present invention.

[0033] In this embodiment, the envelope signal is segmented, and the length of each segment is half the sampling frequency. The slope of each segment is calculated, and segments with similar slopes form a sub-signal. The signal under nonlinear speed change conditions can be divided into multiple sub-signals with linear speed changes. The slope is calculated as follows: ; Where, Indicates that the length of each small segment of the signal is half the sampling frequency, Indicates that the original envelope signal is divided into part, Indicates the The mean value of the time axis of the segment envelope signal, Indicates the The mean value of the segment envelope signal amplitude, Indicates that the length of each segment is W The time variable of the envelope signal, Indicates the amplitude of each small segment of the envelope signal, Indicates slope calculation. Figure 4 As shown in FIG, the envelope signal is divided into four sub-signals according to the distribution of the slope value of each small segment.

[0034] Step S3: perform frequency domain normalization on each sub-signal.

[0035] Figure 5 It is a time-frequency representation diagram of the signal after synchronous time distortion in the present invention.

[0036] In this embodiment, synchronous time warping based on a phase function is used to adjust the scale of the time axis of the non-stationary vibration signal to achieve non-uniform resampling. The formula is as follows: ; Where, Indicates the k The inverse of the phase function of the sub-signal, Indicates the k The result after non-uniform resampling of the sub-signals is: Indicates the k The amplitude parameters of the sub-signals, represents the correction coefficient, A time variable representing the time-varying signal, Indicates the k The phase function of the sub-signal.

[0037] Then, by finding a unified fundamental frequency to align the characteristic frequencies under different speed conditions, the optimization problem is expressed as: ; Where, represents the fundamental frequency, Indicates the k The characteristic ridge line after the sub-signal is resampled, Indicates taking the maximum value, Indicates taking the minimum value. Figure 5 As shown, the characteristic frequencies of different sub-signals have been unified to the same basic frequency.

[0038] Step S4: solving the amplitude normalization function.

[0039] Specifically, in step 4, the amplitude normalization function is solved by power fitting to achieve time domain normalization.

[0040] Figure 6 This is a diagram of the processing result of the normalization function in the time domain in the present invention.

[0041] In this embodiment, time domain normalization is achieved by solving the amplitude normalization function through power fitting, which can effectively eliminate the amplitude modulation effect caused by speed and retain the fault feature information in the vibration signal. The specific formula is as follows: ; ; Where, Indicates the speed signal, represents the normalization function, and represents the linear fitting parameter, represents the power fitting parameter, Indicates the k The result after resampling the sub-signals is Represents the normalized signal. Figure 6 As shown in the figure, the amplitude modulation phenomenon caused by speed fluctuation in the time domain has been eliminated.

[0042] Step S5: Calculate the root mean square characteristics of the normalized signal to diagnose the operating state of the rotating machinery under non-stationary working conditions. Figure 7 It is a root mean square characteristic curve diagram of the normalized signal in the present invention.

[0043] In this embodiment, the root mean square characteristic curve of the normalized signal is calculated to determine whether the current operating state of the rotating machinery is faulty. The root mean square calculation formula is as follows: ; Where, n Indicates the signal length, Represents the normalized signal. For the RMS characteristic curve of the normalized signal, it can be used to determine whether a fault has occurred based on the set alarm threshold. Figure 7As shown in Figure 3, a fixed warning threshold can be used to distinguish between healthy and faulty operating states of rotating machinery under non-stationary conditions.

[0044] Example 2: Based on Example 1, a device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for normalizing a rotating machinery vibration signal is implemented. This device includes electronic devices such as computers, mobile phones, and tablet computers.

[0045] Example 3: On the basis of Example 1, a storage medium stores a computer program. When the computer program is executed by a processor, the method for normalizing a rotating machinery vibration signal is implemented.

[0046] Such storage media include: U disk, mobile hard disk, read-only memory, random access memory, magnetic disk or optical disk, and other media that can store program code.

[0047] It will be understood that in the development of any actual embodiment, as in any engineering or design project, numerous implementation-specific decisions may be made. Such a development effort may be complex and time-consuming, but for those of ordinary skill having the benefit of this disclosure, the development effort will be a routine task of design, fabrication, and production without undue experimentation.

[0048] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for normalizing a rotating machinery vibration signal, characterized in that: The following steps are involved: Step S1, performing Hilbert transform on the collected non-stationary vibration signal to obtain an envelope signal; Step S2: based on the slope, dividing the envelope signal into a plurality of sub-signals with linearly varying speeds; Step S3: performing frequency domain normalization on each sub-signal; Step S4, solving the amplitude normalization function; Step S5: Calculate the root mean square characteristics of the normalized signal to diagnose the operating state of the rotating machinery under non-stationary working conditions.

2. The method for normalizing a rotating machinery vibration signal according to claim 1, wherein: In step S1, the original non-stationary vibration signal is subjected to Hilbert transform to obtain the envelope signal, and the calculation formula is as follows: ； Where: represents the original vibration signal collected; τ is the integration variable, which is used to represent the value point of the signal x(τ) during the integration process; Represents the envelope signal.

3. The method for normalizing a rotating machinery vibration signal according to claim 1, wherein: In step S2, the envelope signal is segmented. The length of each segment is half the sampling frequency. The slope of each segment is calculated. Segments with similar slopes form a sub-signal. The signal under nonlinear speed change conditions is divided into multiple sub-signals with linear speed changes. The slope calculation formula is as follows: ； Where: Indicates that the length of each small segment of the signal is half the sampling frequency, Indicates that the original envelope signal is divided into part, Indicates the The mean value of the time axis of the segment envelope signal, Indicates the The mean value of the segment envelope signal amplitude, Indicates that the length of each segment is W The time variable of the envelope signal, Indicates the amplitude of each small segment of the envelope signal, Indicates slope calculation.

4. The method for normalizing a rotating machinery vibration signal according to claim 1, wherein: In step 3, the sub-signals are normalized in the frequency domain using synchronized time warping.

5. The method for normalizing a rotating machinery vibration signal according to claim 4, wherein: The synchronous time warping based on the phase function is used to adjust the scale of the time axis of the non-stationary vibration signal to achieve non-uniform resampling. The formula is as follows: ； Where, Indicates the k The inverse of the phase function of the sub-signal, Indicates the k The result after non-uniform resampling of the sub-signals is: Indicates the k The amplitude parameters of the sub-signals, represents the correction coefficient, A time variable representing the time-varying signal, Indicates the k The phase function of the sub-signal; Then, by finding a unified fundamental frequency, the characteristic frequencies under different speed conditions are aligned. The formula is as follows: ； Where, represents the fundamental frequency, Indicates the k The characteristic ridge line after the sub-signal is resampled, Indicates taking the maximum value, Indicates taking the minimum value.

6. The method for normalizing a rotating machinery vibration signal according to claim 1, wherein: In step 4, the amplitude normalization function is solved by power fitting to achieve time domain normalization.

7. A rotating machinery vibration signal normalization method according to claim 6, characterized in that: The amplitude normalization function is solved by power fitting, and the formula for time domain normalization is as follows: ； ； Where: Indicates the speed signal, represents the normalization function, and represents the linear fitting parameter, represents the power fitting parameter, Indicates the k The result after resampling the sub-signals is represents the normalized signal.

8. The method for normalizing a rotating machinery vibration signal according to claim 1, wherein: In step 5, the root mean square characteristic curve of the normalized signal is calculated to determine whether the current operating state of the rotating machinery is faulty. The root mean square calculation formula is as follows: ； Where: n Indicates the signal length, represents the normalized signal; For the root mean square characteristic curve of the normalized signal, it is determined whether a fault occurs according to the set alarm threshold.

9. A device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for normalizing a rotating machinery vibration signal according to any one of claims 1 to 8 is implemented.

10. A storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the rotating machinery vibration signal normalization method according to any one of claims 1 to 8 is implemented.

Images