A gearbox bearing fault diagnosis analysis method based on envelope analysis of spectral kurtosis map

Abstract

The application provides a reducer bearing fault diagnosis analysis method based on envelope analysis of a spectrum kurtosis diagram, and comprises the following steps: acquiring an original signal of a reducer bearing in a working state through a sensor; drawing a spectrum kurtosis diagram of the original signal based on the original signal; calculating a center frequency and a bandwidth based on the spectrum kurtosis diagram; carrying out denoising on the original signal based on the center frequency and the bandwidth to obtain a denoised signal; drawing an envelope diagram of the denoised signal by using an envelope analysis method; and detecting the working state of the reducer bearing based on the envelope diagram to obtain a fault diagnosis result.

Description

Technical Field

[0001] This invention belongs to the field of chemical machinery production, and specifically relates to a method for fault diagnosis and analysis of reducer bearings based on envelope analysis of spectral kurtosis diagrams. Background Technology

[0002] Bearings play a crucial role in the operation of large chemical machinery, especially in speed reducers. Bearings operate in a closed environment, and issues such as high corrosion and high temperatures make it difficult for workers to effectively monitor them during operation. Because real-time health monitoring is impossible, bearing failure can lead to slowdown malfunctions, potentially endangering worker safety and impacting the company's economic benefits. Therefore, real-time health monitoring of speed reducer bearings is essential, and research into bearing fault diagnosis is of great significance in extending bearing lifespan.

[0003] With the continuous changes in the mechanical production environment, various extreme environments and chemical raw materials pose great challenges to reducer bearings. The signals received by sensors contain various noises, and workers alone cannot monitor the health status of reducer bearings in real time. Existing reducer fault diagnosis methods cannot quickly denoise the collected signals, and traditional denoising methods cannot accurately and effectively determine faults. The shortcomings in the accuracy and precision of fault diagnosis are becoming increasingly apparent. Therefore, there is a great deal of room for improvement in the accuracy and speed of fault diagnosis. Summary of the Invention

[0004] Purpose of the invention: To address the problems of low accuracy and poor precision in fault diagnosis and difficulty in real-time monitoring of bearings in existing gearbox fault diagnosis methods, this invention proposes a gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams. Through data processing and noise reduction, it enables real-time monitoring of the health status of gearbox bearings.

[0005] Technical solution: A method for fault diagnosis and analysis of gearbox bearings based on envelope analysis of spectral kurtosis plots, comprising the following steps:

[0006] The original signals of the reducer bearing under working conditions are obtained through sensors;

[0007] Based on the original signal, draw its spectral kurtosis diagram; based on the spectral kurtosis diagram, calculate the center frequency and bandwidth of the original signal;

[0008] Based on the center frequency and bandwidth of the original signal, the original signal is denoised to obtain the denoised signal; the envelope analysis method is used to draw the envelope spectrum of the denoised signal.

[0009] Based on the envelope map, the working condition of the reducer bearing is detected, and its fault diagnosis results are obtained.

[0010] Furthermore, the process of drawing a spectral kurtosis diagram based on the original signal specifically includes:

[0011] The tree filter bank is set to have N layers. The tree filter bank includes a two-segment filter and a three-segment filter. The first layer uses a two-segment tree filter to decompose the original signal into a binary tree structure signal. From the second layer to the 2*N-1th layer, a three-segment tree filter is used to decompose the signal into a 1 / 3 tree structure signal.

[0012] The kurtosis value for each layer within its frequency range can be calculated using the following formula:

[0013]

[0014] In the formula, k is the kurtosis value, x is the mean of the data to be calculated, n is the number of data to be calculated, and υ is the standard deviation of the data to be calculated;

[0015] Based on the calculated kurtosis values, a spectral kurtosis map is plotted.

[0016] Furthermore, the calculation of the center frequency and bandwidth based on the spectral kurtosis map specifically includes:

[0017] Find the layer number containing the maximum kurtosis value and the frequency range of that layer from the spectral kurtosis plot;

[0018] The center frequency CF and bandwidth DW of the original signal are calculated according to the following formula;

[0019] CF = SF * (i + 2) i )*2 -j-1 (3)

[0020] DW = SF * 2 -j-1 (4)

[0021] Where j is the layer number where the maximum kurtosis value is located, i is the i-th filter where the maximum kurtosis value is located in the window, and SF is the sampling frequency.

[0022] Furthermore, the aforementioned method of using envelope analysis to draw the envelope map of the denoised signal specifically includes:

[0023] The envelope signal of the denoised signal is calculated using the Fast Fourier Transform method.

[0024] The envelope spectrum of the envelope signal is calculated using the Hilbert transform method.

[0025] The envelope spectrum is plotted based on the envelope spectrum.

[0026] Furthermore, the Fast Fourier Transform method is expressed as follows:

[0027]

[0028] In the formula, w represents frequency, t represents time, and e represents frequency. -iwt It is a complex function.

[0029] Furthermore, the Hilbert transform method is expressed as:

[0030]

[0031] Furthermore, the aforementioned detection of the working state of the reducer bearing based on the envelope map to obtain its fault diagnosis results specifically includes:

[0032] The frequencies presented in the obtained envelope spectrum are compared with the fault frequencies to obtain the fault diagnosis results of the reducer bearing.

[0033] The fault frequency is the fault frequency presented by the envelope spectrum of a signal with a known fault type.

[0034] This invention also discloses a system for a gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams. The system includes a network interface, a memory, and a processor; wherein...

[0035] The network interface is used for receiving and sending signals during the process of sending and receiving information with other external network elements;

[0036] The memory is used to store computer program instructions that can run on the processor;

[0037] The processor is configured to execute, when running the computer program instructions, the steps of the above-disclosed method for fault diagnosis and analysis of gearbox bearings based on envelope analysis of spectral kurtosis diagrams.

[0038] The present invention also discloses a computer storage medium storing a program for a gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams. When the program for the gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams is executed by at least one processor, it implements the steps of the gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams disclosed above.

[0039] Beneficial effects: This invention has the following advantages:

[0040] (1) The method of the present invention solves the problem that the existing high-noise environment cannot perform real-time noise reduction of the collected signal to monitor the real-time health status.

[0041] (2) The method of the present invention can quickly find the center frequency and bandwidth of the signal, thereby denoising the signal and improving the ability to monitor the health status of the bearing in real time. Attached image description:

[0042] Figure 1 This is a flowchart illustrating a method for fault diagnosis and analysis of gearbox bearings based on envelope analysis of spectral kurtosis plots.

[0043] Figure 2 Kurtosis plots of bearing signal spectra under different loads;

[0044] Figure 3 Kurtosis plots of bearing signal spectra at different sampling frequencies;

[0045] Figure 4 Envelope spectra of outer ring signal faults under different loads. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further explained below with reference to the accompanying drawings.

[0047] Example 1:

[0048] This embodiment proposes a method for diagnosing gearbox bearing faults based on spectral kurtosis map envelope signals, which specifically includes the following steps:

[0049] Step 1: Acquire the raw signals of the reducer bearing under working conditions using sensors;

[0050] Step 2: Based on the original signal, draw its spectral kurtosis diagram; based on the spectral kurtosis diagram, calculate the center frequency and bandwidth of the original signal;

[0051] Step 3: Based on the center frequency and bandwidth of the original signal, denoise the original signal to obtain the denoised signal; use envelope analysis to draw the envelope spectrum of the denoised signal;

[0052] Step 4: Based on the envelope map, analyze the working status of the reducer bearing, obtain its fault diagnosis results, and monitor the bearing health status in real time.

[0053] Example 2:

[0054] like Figure 1 As shown in the figure, this embodiment proposes a method for diagnosing gearbox bearing faults based on spectral kurtosis map envelope signals, which specifically includes the following steps:

[0055] Step 1: Acquire the raw signals of the reducer bearing under working conditions using sensors;

[0056] Step 2: Based on the original signal, draw its spectral kurtosis diagram; specific steps include:

[0057] S210: Set the number of windows for the spectral kurtosis plot, corresponding to the number of layers in the tree filter bank. In this embodiment, the tree filter bank consists of a two-segment filter and a three-segment filter, used to segment the frequency. The kurtosis value of each frequency segment is calculated by setting the windows. The first layer uses a two-segment tree filter to decompose the original signal into a binary tree structure. From the second layer to the 2*N-1th layer, a three-segment tree filter is used to decompose the signal obtained by the two-segment filter into a 1 / 3 tree structure. In this embodiment, by combining the two-segment filter and the three-segment filter, the number of frequency segmentation layers can be increased, so that the center frequency and bandwidth of the original signal can be found more quickly.

[0058] In this embodiment, the order of the two-segmented tree filter is 16, the cutoff frequency is 0.4, and the obtained high (low) pass filter coefficients are convolved with the signal.

[0059] The formula relating the coefficients of a two-segment high-pass filter to those of a low-pass filter is:

[0060] g = h(1 + mod(1-n, N)) * (-1) 1-n (1)

[0061] This formula calculates the high-pass filter coefficients from the low-pass filter coefficients obtained from the filter, and then convolves the resulting high (low) pass filter coefficients with the signal data. Here, mod() is the remainder function, 1-n is the dividend, N is the divisor, g is the high-pass filter coefficient, h is the low-pass filter coefficient, N is the filter order, and n ranges from 2 to N+1.

[0062] In this embodiment, a three-segment tree filter with an order of 24 and a cutoff frequency of 0.4 is used. The filter coefficients are calculated and then convolved with the signal.

[0063] S220: Kurtosis is a characteristic of a signal in the time domain. A rapid increase in kurtosis value indicates that the probability density of the vibration signal deviates from a normal distribution. This is an important basis for simplified bearing diagnostic techniques.

[0064] This embodiment calculates the kurtosis value of different frequency ranges under different windows using the kurtosis value formula, and then draws a spectral kurtosis map. The spectral kurtosis map can more intuitively show the window where the maximum spectral kurtosis is located and the center frequency. Then, the original signal is denoised using the obtained center frequency.

[0065] The specific formula for kurtosis is as follows:

[0066]

[0067] In the formula, x is the mean of the data to be calculated, n is the number of data points to be calculated, and υ is the standard deviation of the data to be calculated.

[0068] Step 3: Based on the spectral kurtosis map, find the window layer number and frequency range where the maximum kurtosis value is located.

[0069] Step 4: Calculate the center frequency and bandwidth of the original signal by finding the window layer and frequency range where the maximum kurtosis value is located.

[0070] The specific formula for the center frequency is as follows:

[0071] CF = SF * (i + 2) i )*2 -j-1 (3)

[0072] Where j is the number of layers in the window, i is the i-th filter containing the maximum kurtosis value in the window, CF is the center frequency, and SF is the sampling frequency.

[0073] DW = SF * 2 -j-1 (4)

[0074] Where j is the number of layers in the window, i is the i-th filter containing the maximum kurtosis value in the window, CF is the center frequency, and DW is the bandwidth.

[0075] This embodiment calculates the center frequency and bandwidth of the original signal by plotting a spectral kurtosis diagram. See [link to relevant documentation]. Figure 2 and Figure 3 For bearing signals under different conditions, the spectral kurtosis diagrams under different conditions show that the darkest frequency bands vary in length. This is because the higher the window is located, the fewer the frequency bands are divided into, and the larger the frequency range. Therefore, the length of the darkest color in the spectral kurtosis diagrams under different conditions is different.

[0076] The figure shows a clear range of color depth, which represents the kurtosis value of each frequency band. The darkest color is the frequency band where the maximum kurtosis value is located. By observation, it can be found that the frequency band where the maximum kurtosis value is located can be found through the spectral kurtosis diagram for signals under different loads and different sampling frequencies.

[0077] Step 5: Based on the calculated center frequency and bandwidth of the original signal, use a bandwidth filter to filter the original signal to obtain the denoised signal.

[0078] Step 6: Process the denoised signal using envelope analysis and plot its envelope diagram. Specific steps include:

[0079] The envelope analysis method in this embodiment is a combination of Fast Fourier Transform (FFT) and Hilbert Transform (HFT). Specifically, the FFT is used to calculate the envelope signal of the denoised signal. Based on this envelope signal, the Hilbert Transform is used to calculate the envelope spectrum. Finally, the envelope spectrum diagram is plotted.

[0080] The formula for Fast Fourier Transform is shown below:

[0081]

[0082] Where w represents frequency, t represents time, and e -iwt Since it is a complex function, the purpose of the Fast Fourier Transform is to transform a signal in the time domain into a signal in the frequency domain.

[0083] The Hilbert transform formula is shown below:

[0084]

[0085] Step 7: Based on the envelope map, the working condition of the reducer bearing is detected, specifically as follows:

[0086] The envelope plot will show the frequency of the denoised signal. By analyzing and comparing this frequency with the fault frequency shown in the envelope plot of a signal with a known fault type, the fault diagnosis result of the gearbox bearing can be obtained. (See also...) Figure 4 The dashed and vertical lines represent the outer ring fault frequency and its harmonics. By comparing the fault frequency and its harmonics presented in the envelope spectrum obtained in step 7 with these, it is found that the fault diagnosis result of the reducer bearing is an outer ring fault.

[0087] Example 3:

[0088] This embodiment proposes a gearbox bearing fault diagnosis and analysis system (i.e., a computer device) based on envelope analysis of spectral kurtosis diagrams. The system includes a network interface, a memory, and a processor. The network interface is used for receiving and sending signals during information exchange with other external network elements. The memory is used to store computer program instructions that can be run on the processor. The processor is used to execute the steps of the gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams disclosed above when running the computer program instructions.

[0089] The gearbox bearing fault diagnosis and analysis system (i.e., computer equipment) based on envelope analysis using spectral kurtosis diagrams includes a processor, memory, and network interface connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores data. The network interface allows for communication with external terminals via a network connection.

[0090] Example 4:

[0091] This embodiment proposes a computer storage medium storing a program for a gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams. When the program for the gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams is executed by at least one processor, it implements the steps of the disclosed gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams.

[0092] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0093] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0094] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for fault diagnosis and analysis of gearbox bearings based on envelope analysis of spectral kurtosis plots, characterized in that: Includes the following steps: The original signals of the reducer bearing under working conditions are obtained through sensors; Based on the original signal, draw its spectral kurtosis diagram; based on the spectral kurtosis diagram, calculate the center frequency and bandwidth of the original signal; Based on the center frequency and bandwidth of the original signal, the original signal is denoised to obtain the denoised signal. The envelope analysis method is used to draw the envelope map of the denoised signal; Based on the envelope map, the working condition of the reducer bearing is detected, and its fault diagnosis results are obtained; Specifically, drawing the spectral kurtosis diagram based on the original signal includes: The number of windows N in the spectral kurtosis plot is set, and the number of windows corresponds to the number of layers in the tree filter bank; the tree filter bank consists of a two-segment tree filter and a three-segment tree filter, which are used to segment the frequency; the kurtosis value of each frequency segment is calculated by setting the windows; The first layer uses a two-segment tree filter to decompose the original signal into a binary tree structure signal. From the second layer to the 2*N-1th layer, a three-segment tree filter is used to decompose the signal obtained by the two-segment filter into a 1 / 3 tree structure signal. The two-segmented tree filter has an order of 16 and a cutoff frequency of 0.

4. The resulting high-pass filter coefficients and low-pass filter coefficients are convolved with the signal. The formula relating the high-pass filter coefficients to the low-pass filter coefficients of a two-segmented tree filter is: g=h(1+mod(1-n,N))*(-1) 1-n (1) Where mod() is the remainder function, 1-n is the dividend, N is the divisor, g is the high-pass filter coefficient, h is the low-pass filter coefficient, N is the order of the filter, and n takes the value from 2 to N+1. A three-segment tree filter with order 24 and a cutoff frequency of 0.

4. The kurtosis values ​​for different frequency ranges within different windows were calculated using the kurtosis value formula: In the formula, k is the kurtosis value, x is the mean of the data to be calculated, n is the number of data to be calculated, and υ is the standard deviation of the data to be calculated; Based on the calculated kurtosis values, a spectral kurtosis map is plotted.

2. The method for fault diagnosis and analysis of reducer bearings based on envelope analysis of spectral kurtosis diagrams according to claim 1, characterized in that: The calculation of the center frequency and bandwidth of the original signal based on the spectral kurtosis diagram specifically includes: Find the layer number containing the maximum kurtosis value and the frequency range of that layer from the spectral kurtosis plot; The center frequency CF and bandwidth DW of the original signal are calculated according to the following formula; CF=SF*(i+2 i )*2 -j-1 (3)DW=SF*2 -j-1 (4) Where j is the layer number where the maximum kurtosis value is located, i is the i-th filter where the maximum kurtosis value is located in the window, and SF is the sampling frequency.

3. The method for fault diagnosis and analysis of reducer bearings based on envelope analysis of spectral kurtosis diagrams according to claim 1, characterized in that: The aforementioned method of using envelope analysis to draw the envelope map of the denoised signal specifically includes: The envelope signal of the denoised signal is calculated using the Fast Fourier Transform method. The envelope spectrum of the envelope signal is calculated using the Hilbert transform method. The envelope spectrum is plotted based on the envelope spectrum.

4. The method for fault diagnosis and analysis of reducer bearings based on envelope analysis of spectral kurtosis plots according to claim 3, characterized in that: The Fast Fourier Transform method is expressed as follows: In the formula, w represents frequency, t represents time, and e represents frequency. -iwt It is a complex function.

5. The method for fault diagnosis and analysis of reducer bearings based on envelope analysis of spectral kurtosis plots according to claim 3, characterized in that: The Hilbert transform method is expressed as follows:

6. The method for fault diagnosis and analysis of reducer bearings based on envelope analysis of spectral kurtosis diagrams according to claim 1, characterized in that: The aforementioned method of detecting the operating status of the reducer bearing based on envelope maps and obtaining its fault diagnosis results specifically includes: The frequencies presented in the obtained envelope spectrum are compared with the fault frequencies to obtain the fault diagnosis results of the reducer bearing. The fault frequency is the fault frequency presented by the envelope spectrum of a signal with a known fault type.

7. A system for a gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis plots, characterized in that, The system includes a network interface, memory, and a processor; wherein... The network interface is used for receiving and sending signals during the process of sending and receiving information with other external network elements; The memory is used to store computer program instructions that can run on the processor; The processor is configured to, when running the computer program instructions, execute the steps of the gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams as described in any one of claims 1 to 6.

8. A computer storage medium, characterized in that, The computer storage medium stores a program for a gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams. When the program for the gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams is executed by at least one processor, it implements the steps of the gearbox bearing fault diagnosis and analysis method based on envelope analysis of spectral kurtosis diagrams as described in any one of claims 1 to 6.

Images