Explosion-proof motor bearing fault detection method based on frequency domain sparse variational mode extraction

Abstract

The present invention provides an explosion-proof motor bearing fault detection method based on frequency domain sparse variational mode extraction, comprising the following steps: S1: using a vibration sensor to collect the vibration signal of the device under test, then intercepting the vibration signal per unit time and removing the mean; S2: constructing a frequency domain sparse variational mode extraction model, deriving an iterative solution algorithm for the model, and then extracting fault features from the vibration signal through the iterative solution algorithm; S3: analyzing the time domain waveform and envelope spectrum of the fault features, calculating the theoretical fault characteristic frequency of the bearing according to the bearing structural parameters and the rotational speed, comparing each spectrum line with the theoretical fault characteristic frequency of the bearing, and determining the bearing fault type according to the matching results; the present invention can still accurately extract weak fault features under multiple interference components, realize accurate detection of explosion-proof motor bearing faults, and ensure the safe operation of the entire mechanical system.

Description

Technical Field

[0001] The present invention relates to the technical field of bearing fault detection, and in particular to an explosion-proof motor bearing fault detection method based on frequency domain sparse variational mode extraction. Background Art

[0002] Explosion-proof motor bearings are critical components in motors. Their complex operating environment makes them prone to failure. Bearing failures can cause motor failure, economic losses, and even casualties. Therefore, developing advanced bearing fault detection technologies to ensure motor reliability is crucial. When localized faults such as pitting, spalling, or cracking occur in explosion-proof motor bearings, periodic transients appear in the vibration signal. These periodic transients are characteristic of bearing failures and are key indicators for fault detection. However, in practical applications, these fault signatures are often too subtle to be easily overwhelmed by interference from the original vibration signal, posing a challenge to fault detection.

[0003] Signal decomposition methods have been a research hotspot in signal processing in recent years and have also found widespread application in fault detection. Using demodulation techniques, they adaptively decompose multimodal signals into multiple modes and simultaneously extract the frequencies corresponding to each mode. Therefore, signal decomposition methods offer a viable approach to fault feature extraction.

[0004] Variational mode extraction (VME) is a novel signal decomposition method proposed in recent years and widely used in signal processing. It builds on VME (which combines demodulation techniques with Wiener filters to decompose multimodal signals) by adding new criteria to reduce modal aliasing, thereby extracting the desired mode. This method considers the spectral overlap between the desired mode and the interfering components in the signal, making it ideal for extracting transient characteristics caused by bearing failures in explosion-proof motors.

[0005] However, the traditional variational mode extraction algorithm does not take into account the frequency domain sparsity of fault characteristics. At the same time, due to the complex working environment of explosion-proof motor bearings and the limitations of machining accuracy and equipment installation accuracy, even when the explosion-proof motor bearings are in a healthy state, there is inevitably interference from noise and harmonic components in the vibration signal. These factors greatly reduce the performance of the variational mode extraction algorithm in fault feature extraction, which may lead to detection errors in explosion-proof motor bearing fault detection.

[0006] Therefore, a new solution is urgently needed to solve the defects and shortcomings in the above-mentioned prior art. Summary of the Invention

[0007] In order to solve the defects and shortcomings in the prior art, the present invention provides an explosion-proof motor bearing fault detection method based on frequency domain sparse variational mode extraction.

[0008] The specific solution provided by the present invention is:

[0009] The method for detecting bearing faults of explosion-proof motors based on frequency domain sparse variational mode extraction is characterized by comprising the following steps:

[0010] S1: Use the vibration sensor to collect the vibration signal of the device under test, then intercept the vibration signal per unit time and remove the mean value, and record the vibration signal as ;

[0011] S2: Construct a frequency domain sparse variational mode extraction model, derive the iterative solution algorithm of the model, and then use the iterative solution algorithm to extract the vibration signal. Extract fault features from

[0012] S3: Analyze the time domain waveform and envelope spectrum of the fault characteristics, calculate the theoretical bearing fault characteristic frequency based on the bearing structural parameters and speed, compare each spectrum line with the theoretical bearing fault characteristic frequency, and determine the bearing fault type based on the matching results.

[0013] As a further preferred embodiment of the present invention, the constructed frequency domain sparse variational mode extraction model satisfies:

[0014] ;

[0015] And satisfy: ;

[0016] Where:

[0017] The first term represents an indication of the modal bandwidth, where is the expected mode, Is calculated relative to time The partial derivative of , that is, the Wiener filter; is the Dirac distribution, defined as t =0 is equal to zero at points other than , and the integral over the entire domain is equal to 1; * indicates convolution, is the frequency corresponding to the desired mode, represents the regularization parameter, represents an imaginary number, Indicates time;

[0018] The second term represents the spectrum overlap indication, where is the residual signal after extracting the desired mode, is the filter impulse response function;

[0019] The third term is the frequency domain sparse constraint regularization term, where is the regularization parameter, is the element-wise multiplication operation, A represents the Fourier transform matrix, Represents frequency domain sparse coefficients No. groups, which contain elements; is a non-convex penalty function, and It is a periodic vector and weighted weight customized according to the frequency domain sparsity of fault characteristics.

[0020] As a further preferred embodiment of the present invention, in the frequency domain sparse variational mode extraction model,

[0021] The period vector Defined as:

[0022] ;

[0023] Where: is the number of non-zero points, corresponding to the number of large amplitude points in the fault transient; The number of zero points is shown, and the corresponding frequency interval is , represents the number of small amplitude points between adjacent fault transients; S means that the group contains elements; P represents b The number of fault transients in .

[0024] As a further preferred embodiment of the present invention, in the frequency domain sparse variational mode extraction model,

[0025] Introducing the MCRn criterion to determine the weighted weight w n , for the Group, defines the standard MCRn as:

[0026] ;

[0027] In the formula It is a vibration signal Sparse coefficients in the frequency domain, binary vector , Defined as:

[0028] ;

[0029] Using standard MCRn, define The weights of the groups are:

[0030] ;

[0031] Where “1” indicates that the fault transient sparse coefficient is given a high weight so that it can be retained; " indicates that the sparse coefficients of the harmonic components are given low weights to suppress them; Indicates the threshold value.

[0032] As a further preferred embodiment of the present invention, in the frequency domain sparse variational mode extraction model,

[0033] Non-convex penalty function The convexity of Determine and make the parameters satisfy:

[0034] .

[0035] As a further preferred embodiment of the present invention, in step S2, deriving the iterative solution algorithm of the model includes the following steps:

[0036] S21: Input parameters , , , , , , ,in Indicates update parameters;

[0037] S22: Initialization and , the number of iterations The initial number of iterations is set to 0, that is, ; and Estimate the initial frequency, i.e. Initial estimate; where represents the mth mode at the first iteration; represents the Lagrange multiplier at the first iteration;

[0038] S23: The number of iterations starts from 0 and iterates until n+1 times, that is ;

[0039] S24: Find the optimal solution for the non-convex penalty function using the following formula:

[0040] ;

[0041] in, represents the derivative of the non-convex penalty function; j Indicates the n In the group j elements;

[0042] S25: Calculate weighted weights and construct the weighted matrix ;

[0043] S26: Update according to the following formula And ensure that all :

[0044] ;

[0045] in, m Indicates the number of modes; represents the Lagrange multiplier; Represents an optimizer for a non-convex penalty function;

[0046] S27: Update according to the following formula :

[0047] ;

[0048] S28: Update according to the following formula And ensure all :

[0049] ;

[0050] S29: Determine whether the convergence condition is met, namely:

[0051] ;

[0052] in, represents the convergence index;

[0053] S210: When the judgment result is that the convergence condition is met, output ;

[0054] in, is the expected mode, that is, the fault feature extracted from the signal with multiple interference components.

[0055] As a further preferred embodiment of the present invention, in step S3, when calculating the theoretical bearing fault characteristic frequency based on the bearing structural parameters and the rotational speed, the theoretical bearing fault characteristic frequency is calculated according to the following formula:

[0056] ;

[0057] ;

[0058] ;

[0059] in,

[0060] 、 and Represent the fault characteristic frequencies of the bearing outer ring, bearing inner ring and bearing roller respectively;

[0061] Indicates the rotation frequency of the bearing;

[0062] Indicates the number of rollers; Indicates the bearing raceway diameter; Indicates roller diameter;

[0063] Represents the contact angle.

[0064] As a further preferred embodiment of the present invention, in step S3, when comparing each spectral line with the theoretical bearing fault characteristic frequency and determining the bearing fault type according to the matching result, the following steps are included:

[0065] S31: Convert the fault features extracted from the vibration signal into envelope spectrum.

[0066] S32: Compare the spectrum lines in the envelope spectrum with the theoretical fault characteristic frequencies of the bearing. If they are consistent, it is determined that a fault exists at the current position of the bearing; otherwise, it is determined that no fault exists at the current position of the bearing.

[0067] Compared with the existing technology, the present invention can achieve the following technical effects:

[0068] 1) The present invention provides an explosion-proof motor bearing fault detection method based on frequency-domain sparse variational mode extraction, which fully considers the frequency-domain sparsity of fault characteristics. In the frequency domain, periodicity is first used to promote the sparsity of the periodic group of fault characteristics. Secondly, a weighted strategy is used to hierarchically penalize the sparse coefficients of fault characteristics and harmonic components. Finally, a non-convex penalty function is used to further highlight the frequency-domain sparsity of fault characteristics, ultimately forming a frequency-domain sparse constraint to achieve the purpose of enhancing weak fault characteristics. Therefore, even in the presence of multiple interference components, weak fault characteristics can still be accurately extracted, achieving accurate detection of explosion-proof motor bearing faults and ensuring the safe operation of the entire mechanical system. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 A flowchart of the steps of the fault detection method provided by the present invention;

[0070] Figure 2 It is the time domain waveform of the vibration signal;

[0071] Figure 3 is the envelope spectrum of the vibration signal;

[0072] Figure 4 A time domain waveform diagram of features extracted by the method of the present invention;

[0073] Figure 5 The envelope spectrum of the feature extraction method of the present invention;

[0074] Figure 6Time domain waveform diagram for extracting features using the variational mode extraction method;

[0075] Figure 7 Envelope spectrum of features extracted by variational mode extraction method. DETAILED DESCRIPTION

[0076] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0077] In the description of the present invention, it should be noted that the terms "upper," "lower," "inner," "outer," "front end," "rear end," "both ends," "one end," "the other end," and the like, indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and therefore should not be construed as limiting the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0078] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "installed," "provided with," "connected," etc., should be understood in a broad sense. For example, "connected" may refer to a fixed connection, a detachable connection, or an integral connection; it may refer to a mechanical connection or an electrical connection; it may refer to a direct connection or an indirect connection through an intermediate medium; it may refer to internal communication between two components. Those skilled in the art will be able to understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0079] [First embodiment]

[0080] like Figure 1 The first embodiment of the present invention is shown, which provides a method for detecting bearing faults of explosion-proof motors based on frequency-domain sparse variational mode extraction, comprising the following steps:

[0081] S1: Install the vibration sensor on the device under test, use the vibration sensor and data acquisition card to collect the vibration signal of the device under test, then intercept the vibration signal per unit time (for example, 1s) and remove the mean. The data distribution is made more balanced by the data processing method of removing the mean, thereby improving the efficiency and effect of model training. The vibration signal is recorded as , as shown in Figure 2. However, no obvious periodic fault transients were found. Figure 3 shows the envelope spectrum of the vibration signal, from which only the rotation frequency was identified. No bearing fault characteristic frequency can be identified , that is, the envelope spectrum does not provide any fault information, so the bearing fault cannot be detected.

[0082] S2: Construct a frequency domain sparse variational mode extraction model, use the alternating direction multiplier method and controlled minimization method to derive the iterative solution algorithm of the model, and then use the iterative solution algorithm to extract the vibration signal. Extract fault features from

[0083] In this embodiment, the constructed frequency domain sparse variational mode extraction model satisfies:

[0084] ;

[0085] And satisfy: ;

[0086] By using this model to find the objective function (i.e. The function within produces the minimum value 、 and ;

[0087] Where: The first term indicates the modal bandwidth, is the expected mode, Is calculated relative to time The partial derivative of the Wiener filter is used to transform the desired mode into After demodulation to baseband through demodulation technology, it is combined with Wiener filter for smoothing; is the Dirac distribution, defined as t =0 is equal to zero at points other than , and the integral over the entire domain is equal to 1; * indicates convolution, is the frequency corresponding to the desired mode, represents the regularization parameter, represents an imaginary number, Indicates time;

[0088] The second term indicates the spectrum overlap indication, which aims to make the residual signal after extracting the desired mode There should be no energy or less energy at the frequency of the desired mode to solve the problem of modal aliasing; is the residual signal after extracting the desired mode, is the filter impulse response function;

[0089] The third term is the frequency domain sparse constraint regularization term, which takes into account the frequency domain sparsity of the desired mode; is the regularization parameter, is the element-wise multiplication operation, A represents the Fourier transform matrix, Represents frequency domain sparse coefficients No. groups, which contain elements; is a non-convex penalty function, and It is a periodic vector and weighted weight customized according to the frequency domain sparsity of fault characteristics.

[0090] In the constructed frequency domain sparse variational mode extraction model, for the periodic vector , which is defined as:

[0091] ;

[0092] Where: is the number of non-zero points, corresponding to the number of large amplitude points in the fault transient; The number of zero points is shown, and the corresponding frequency interval is The number of small amplitude points between adjacent fault transients; S means that the group contains elements; P represents b In this embodiment, points with large amplitudes are marked as 1, and points with small amplitudes are marked as 0 in the matrix.

[0093] For weighted weights w n

[0094] The MCRn criterion is introduced to determine the weighted weights, and an MCRn criterion for judging the maximum contribution rate of the fault transient and harmonic component frequency lines is introduced to determine the weighting strategy:

[0095] For the Group, defines the standard MCRn as:

[0096] ;

[0097] In the formula It is a vibration signal Sparse coefficients in the frequency domain, binary vector , Defined as:

[0098] ;

[0099] Using standard MCRn, define The weights of the groups are:

[0100] ;

[0101] Where “1” indicates that the fault transient sparse coefficient is given a high weight so that it can be retained; " indicates that the sparse coefficients of the harmonic components are given low weights to suppress them; Indicates the threshold; in this embodiment, a high weight indicates a larger weight, and a low weight indicates a smaller weight. The threshold for distinguishing the two can be pre-set and adjusted in real time according to actual conditions. In this embodiment, a high weight is recorded as 1, and a low weight is recorded as .

[0102] For non-convex penalty functions The convexity of To ensure that the objective function of the optimization problem (1) is strictly convex, it is necessary to make the parameters satisfy:

[0103] .

[0104] In step S2, for solving the variational model, under the framework of the desired mode and frequency alternating update strategy, an iterative solution algorithm is derived using the alternating direction multiplier method and the controlled minimization method, and the following iterative solution algorithm for the frequency domain sparse variational mode extraction model is obtained. That is, the iterative solution algorithm for the model is derived, which includes the following steps:

[0105] S21: Input parameters , , , , , , ,in Indicates update parameters;

[0106] S22: Initialization and , the number of iterations The initial number of iterations is set to 0, that is, ; and Estimate the initial frequency, i.e. Initial estimate; where represents the mth mode at the first iteration; represents the Lagrange multiplier at the first iteration;

[0107] S23: The number of iterations starts from 0 and iterates until n+1 times, that is ;

[0108] S24: Find the optimal solution for the non-convex penalty function using the following formula:

[0109] ;

[0110] in, represents the derivative of the non-convex penalty function; j Indicates the n In the group j elements;

[0111] S25: Calculate weighted weights and construct the weighted matrix ;

[0112] S26: Update according to the following formula And ensure all :

[0113] ;

[0114] in, m Indicates the number of modes; represents the Lagrange multiplier; Represents an optimizer for a non-convex penalty function;

[0115] S27: Update according to the following formula :

[0116] ;

[0117] S28: Update according to the following formula And ensure all :

[0118] ;

[0119] S29: Determine whether the convergence condition is met, namely:

[0120] ;

[0121] in, represents the convergence index;

[0122] S210: When the judgment result is that the convergence condition is met, output ;

[0123] in, is the expected mode, that is, the fault feature extracted from the signal with multiple interference components.

[0124] S3: Analyze the time domain waveform and envelope spectrum of the fault characteristics, calculate the theoretical bearing fault characteristic frequency based on the bearing structural parameters and speed, compare each spectrum line with the theoretical bearing fault characteristic frequency, and determine the bearing fault type based on the matching results;

[0125] When calculating the theoretical bearing fault characteristic frequency based on the bearing structural parameters and speed, the following formula is used to calculate the theoretical bearing fault characteristic frequency:

[0126] ;

[0127] ;

[0128] ;

[0129] in,

[0130] 、 and Represent the fault characteristic frequencies of the bearing outer ring, bearing inner ring and bearing roller respectively;

[0131] Indicates the rotation frequency of the bearing;

[0132] Indicates the number of rollers; Indicates the bearing raceway diameter; Indicates roller diameter;

[0133] Represents the contact angle.

[0134] When comparing each spectral line with the theoretical bearing fault characteristic frequency and determining the bearing fault type based on the matching results, the following steps are included:

[0135] S31: converting the fault features extracted from the vibration signal into an envelope spectrum;

[0136] S32: Compare the spectrum lines in the envelope spectrum with the theoretical fault characteristic frequencies of the bearing. If they are consistent, it is determined that a fault exists at the current position of the bearing; otherwise, it is determined that no fault exists at the current position of the bearing.

[0137] The time domain waveform and envelope spectrum of the fault characteristics extracted by the method of the present invention in the embodiment are respectively as follows: Figure 4 and Figure 5 As shown. Figure 4 In the , clear periodic fault characteristics can be observed, indicating that the serious interference components are effectively removed. Figure 5 In the envelope spectrum, the outer race fault characteristic frequency The frequency and its multiples are also very significant. In addition, since the outer ring of the bearing is driven by the drive motor, clear sidebands can be observed in the envelope spectrum. These fault information shows that the locomotive bearing outer ring fault has been accurately detected.

[0138] The time domain waveform and envelope spectrum of the features extracted by the traditional variational mode extraction method are as follows: Figure 6 and Figure 7As shown in the time domain waveform, no obvious periodic transients are found, and the envelope spectrum can barely be identified. and its sidebands, but due to the influence of multiple interference components, it is impossible to determine whether it is the fault characteristic frequency caused by the outer ring fault, which shows that the variational mode extraction algorithm fails to accurately detect the bearing outer ring fault.

[0139] The method of the present invention can also be set as rolling bearing fault detection software to be installed in a host computer, which is connected to a signal acquisition device. The host computer processes the vibration signal in the above steps in real time to detect rolling bearing faults in a timely manner and determine the fault type, thereby ensuring the safe operation of the entire mechanical system.

[0140] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A method for detecting explosion-proof motor bearing faults based on frequency-domain sparse variational mode extraction, characterized by: The following steps are involved: S1: Use the vibration sensor to collect the vibration signal of the device under test, then intercept the vibration signal per unit time and remove the mean value, and record the vibration signal as ; S2: Construct a frequency-domain sparse variational mode extraction model, derive an iterative solution algorithm for the model, and then extract fault features from the vibration signal through the iterative solution algorithm; S3: Analyze the time domain waveform and envelope spectrum of the fault characteristics, calculate the theoretical bearing fault characteristic frequency based on the bearing structural parameters and speed, compare each spectrum line with the theoretical bearing fault characteristic frequency, and determine the bearing fault type based on the matching results; The constructed frequency domain sparse variational mode extraction model satisfies: ； And satisfy: ; Where: The first term represents an indication of the modal bandwidth, where is the expected mode, Is calculated relative to time t The partial derivative of , that is, the Wiener filter; is the Dirac distribution, defined as t =0 is equal to zero at points other than , and the integral over the entire domain is equal to 1; * indicates convolution, is the frequency corresponding to the desired mode, represents the regularization parameter, j represents an imaginary number, t Indicates time; The second term represents the spectrum overlap indication, where is the residual signal after extracting the desired mode, is the filter impulse response function; The third term is the frequency domain sparse constraint regularization term, where is the regularization parameter, is the element-wise multiplication operation, A represents the Fourier transform matrix, Represents frequency domain sparse coefficients No. groups, which contain elements; This group includes elements; is a non-convex penalty function, and It is a periodic vector and weighted weight customized according to the frequency domain sparsity of fault characteristics; In the frequency domain sparse variational mode extraction model, Introducing the MCRn criterion to determine the weighted weight , for the Group, defines the standard MCRn as: ； In the formula It is a vibration signal Sparse coefficients in the frequency domain, a binary vector, , Defined as: ； Using standard MCRn, define The weights of the groups are: ； Where "1" indicates that the fault transient sparse coefficient is given a high weight so that it can be retained; " indicates that the sparse coefficients of the harmonic components are given low weights to suppress them; Indicates the threshold value.

2. The method for detecting explosion-proof motor bearing faults based on frequency domain sparse variational mode extraction according to claim 1 is characterized in that: In the frequency domain sparse variational mode extraction model, The period vector Defined as: ； Where: is the number of non-zero points, corresponding to the number of large amplitude points in the fault transient; Indicates the number of zero points, corresponding to the frequency interval , represents the number of small amplitude points between adjacent fault transients; Indicates that the group contains elements; P represents b The number of fault transients in .

3. The method for detecting explosion-proof motor bearing faults based on frequency domain sparse variational mode extraction according to claim 2, characterized in that: In the frequency domain sparse variational mode extraction model, Non-convex penalty function The convexity of Determine and make the parameters satisfy: 。 4. The method for detecting bearing faults of explosion-proof motors based on frequency-domain sparse variational mode extraction according to claim 3, characterized in that: In step S2, deriving the iterative solution algorithm of the model includes the following steps: S21: Input parameters , , , , , , ,in Indicates update parameters; S22: Initialization and , the number of iterations The initial number of iterations is set to 0, that is, ; and Estimate the initial frequency, i.e. Initial estimate; where represents the mth mode at the first iteration; represents the Lagrange multiplier at the first iteration; S23: The number of iterations starts from 0 and iterates until n+1 times, that is ; S24: Find the optimal solution for the non-convex penalty function using the following formula: ； in, represents the derivative of the non-convex penalty function; j Indicates the n In the group j elements; S25: Calculate weighted weights and construct the weighted matrix ; S26: Update according to the following formula And ensure all : ； in, m Indicates the number of modes; represents the Lagrange multiplier; Represents an optimizer for a non-convex penalty function; S27: Update according to the following formula : ； S28: Update according to the following formula And ensure all : ； S29: Determine whether the convergence condition is met, namely: ； in, represents the convergence index; S210: When the judgment result is that the convergence condition is met, output ; in, is the expected mode, that is, the fault feature extracted from the signal with multiple interference components.

5. The method for detecting bearing faults of explosion-proof motors based on frequency domain sparse variational mode extraction according to claim 1, characterized in that: In step S3, when calculating the theoretical bearing fault characteristic frequency based on the bearing structural parameters and the rotational speed, the theoretical bearing fault characteristic frequency is calculated according to the following formula: ； ； ； in, , ,and Represent the fault characteristic frequencies of the bearing outer ring, bearing inner ring and bearing roller respectively; Indicates the rotation frequency of the bearing; Indicates the number of rollers; Indicates the bearing raceway diameter; Indicates roller diameter; Represents the contact angle.

6. The method for detecting explosion-proof motor bearing faults based on frequency domain sparse variational mode extraction according to claim 1, characterized in that: In step S3, when comparing each spectrum line with the theoretical bearing fault characteristic frequency and determining the bearing fault type based on the matching result, the following steps are included: S31: Convert the fault features extracted from the vibration signal into envelope spectrum. S32: Compare the spectrum lines in the envelope spectrum with the theoretical fault characteristic frequencies of the bearing. If they are consistent, it is determined that a fault exists at the current position of the bearing; otherwise, it is determined that no fault exists at the current position of the bearing.

Images