Bearing fault frequency spectrum evaluation method, device, equipment and medium

Abstract

The invention discloses a bearing fault frequency spectrum evaluation method and device, equipment and a medium, and the method comprises the steps: firstly carrying out the feature extraction of a frequency spectrum of a to-be-evaluated bearing, and obtaining the feature frequency and amplitude of the to-be-evaluated bearing; the theoretical characteristic frequency of the fault of the bearing to be evaluated and the neighborhood of the theoretical characteristic frequency are determined; constructing a distance evaluation matrix by using the Gaussian membership function and the frequencies of the characteristic frequency points of the neighborhood, and constructing a peak value evaluation matrix by using the Gaussian membership function and the amplitudes of the characteristic frequency points of the neighborhood; respectively determining the weight of the distance evaluation matrix and the weight of the peak value evaluation matrix by using a variable coefficient method, and further obtaining fuzzy comprehensive evaluation; and finally, when the fuzzy comprehensive evaluation is greater than a fuzzy comprehensive evaluation health threshold value, determining that the to-be-evaluated bearing has a fault. According to the method, the difference between the theoretical fault characteristic frequency of the bearing and the actually detected frequency spectrum is considered, and the accuracy of evaluating the bearing fault is improved.

Description

Technical Field

[0001] The present application relates to the field of bearing failures, and in particular to a method, device, equipment and medium for evaluating a bearing failure spectrum. Background Art

[0002] Bearings are an indispensable part of machinery, used to support rotating shafts and reduce friction. Almost all equipment with rotating parts requires bearings, such as motors, generators, fans, pumps, etc. They are widely used in rail transportation, automobiles, aerospace, electric power and other industries. For motors, bearings are moving and easily worn parts, and bearing failures account for about 35% of all motor failures. Monitoring the health status of motor bearings helps manage equipment health and increase motor life. Since the general installation position of bearings is inaccessible and non-visible, the vibration signal of the bearing can be detected to determine whether the bearing is faulty.

[0003] At present, there are many algorithms for extracting characteristic signals of bearing faults, but there are few algorithms for subsequent processing of characteristic signals, namely feature selection (spectrum recognition). Commonly used algorithms include principal component analysis (PCA), linear discriminant analysis (LDA), information entropy quantification, etc. for feature selection.

[0004] Due to the rich characteristic signals of bearing faults, such as the vibration acceleration spectrum, stator coil current spectrum, etc., the frequency multiples of their respective spectra may overlap, and the spectra of faults in different parts may also overlap. At this time, the feature selection algorithm may have the problem of inaccurate spectrum identification, which in turn leads to a low accuracy rate in evaluating bearing faults. Summary of the invention

[0005] In view of this, the present application provides a method, device, equipment and medium for evaluating a bearing fault spectrum, aiming to improve the accuracy of evaluating bearing faults.

[0006] In a first aspect, the present application provides a method for evaluating a bearing fault spectrum, the method comprising:

[0007] Extracting features from the frequency spectrum of the bearing to be evaluated to obtain the characteristic frequency and amplitude of the bearing to be evaluated;

[0008] Determine the theoretical characteristic frequency of the bearing to be evaluated and the neighborhood of the theoretical characteristic frequency, wherein the neighborhood is an adjacent area of ​​the theoretical characteristic frequency;

[0009] Constructing a distance evaluation matrix using a Gaussian membership function and the frequencies of the characteristic frequency points of the neighborhood, and constructing a peak evaluation matrix using a Gaussian membership function and the amplitudes of the characteristic frequency points of the neighborhood;

[0010] Determining the weight of the distance evaluation matrix and the weight of the peak evaluation matrix respectively by using a coefficient of variation method;

[0011] Performing fuzzy synthesis on the distance evaluation matrix and the weight of the distance evaluation matrix, and the peak evaluation matrix and the weight of the peak evaluation matrix to obtain a fuzzy comprehensive evaluation;

[0012] If the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a fault.

[0013] Optionally, the types of failures of the bearing to be evaluated include inner ring failure, outer ring failure, cage failure and rolling element failure;

[0014] The step of determining the theoretical characteristic frequency of the bearing failure to be evaluated and the neighborhood of the theoretical characteristic frequency includes:

[0015] Determine the inner ring theoretical characteristic frequency of the inner ring fault of the bearing to be evaluated and the neighborhood of the inner ring theoretical characteristic frequency;

[0016] Determine the outer ring theoretical characteristic frequency of the bearing to be evaluated where an outer ring failure occurs and the neighborhood of the outer ring theoretical characteristic frequency;

[0017] Determine the cage theoretical characteristic frequency of the bearing to be evaluated that has a cage failure and the neighborhood of the cage theoretical characteristic frequency;

[0018] Determine the rolling element theoretical characteristic frequency of the bearing to be evaluated when a rolling element failure occurs and the neighborhood of the rolling element theoretical characteristic frequency.

[0019] Optionally, the constructing a distance evaluation matrix using a Gaussian membership function and the frequencies of the characteristic frequency points of the neighborhood, and the constructing a peak evaluation matrix using a Gaussian membership function and the amplitudes of the characteristic frequency points of the neighborhood, include:

[0020] An inner circle distance evaluation matrix is ​​constructed using the Gaussian membership function and the frequencies of the characteristic frequency points in the neighborhood of the inner circle theoretical characteristic frequency, and an inner circle peak evaluation matrix is ​​constructed using the Gaussian membership function and the amplitudes of the characteristic frequency points in the neighborhood of the inner circle theoretical characteristic frequency;

[0021] An outer circle distance evaluation matrix is ​​constructed using the Gaussian membership function and the frequencies of the characteristic frequency points in the neighborhood of the outer circle theoretical characteristic frequency, and an outer circle peak evaluation matrix is ​​constructed using the Gaussian membership function and the amplitudes of the characteristic frequency points in the neighborhood of the outer circle theoretical characteristic frequency;

[0022] Constructing a cage distance evaluation matrix using the Gaussian membership function and the frequencies of the characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the cage, and constructing a cage peak evaluation matrix using the Gaussian membership function and the amplitudes of the characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the cage;

[0023] A rolling element distance evaluation matrix is ​​constructed using the Gaussian membership function and the frequencies of characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the rolling element, and a rolling element peak evaluation matrix is ​​constructed using the Gaussian membership function and the amplitudes of characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the rolling element.

[0024] Optionally, the using a coefficient of variation method to respectively determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix includes:

[0025] Using the coefficient of variation method, respectively determining the weight of the inner circle distance evaluation matrix and the weight of the inner circle peak value evaluation matrix;

[0026] Using the coefficient of variation method, respectively determining the weight of the outer circle distance evaluation matrix and the weight of the outer circle peak evaluation matrix;

[0027] Using a coefficient of variation method, respectively determining the weight of the cage distance evaluation matrix and the weight of the cage peak evaluation matrix;

[0028] The weight of the rolling element distance evaluation matrix and the weight of the rolling element peak value evaluation matrix are determined respectively by using the coefficient of variation method.

[0029] Optionally, the fuzzy synthesis of the distance evaluation matrix and the weight of the distance evaluation matrix, the peak evaluation matrix and the weight of the peak evaluation matrix to obtain a fuzzy comprehensive evaluation includes:

[0030] Performing fuzzy synthesis on the inner circle distance evaluation matrix and the weight of the inner circle distance evaluation matrix, the inner circle peak evaluation matrix and the weight of the inner circle peak evaluation matrix to obtain an inner circle fuzzy comprehensive evaluation;

[0031] Performing fuzzy synthesis on the outer circle distance evaluation matrix and the weight of the outer circle distance evaluation matrix, the outer circle peak evaluation matrix and the weight of the outer circle peak evaluation matrix to obtain an outer circle fuzzy comprehensive evaluation;

[0032] Performing fuzzy synthesis on the cage distance evaluation matrix and the weight of the cage distance evaluation matrix, the cage peak evaluation matrix and the weight of the cage peak evaluation matrix to obtain a cage fuzzy comprehensive evaluation;

[0033] The rolling body distance evaluation matrix and the weight of the rolling body distance evaluation matrix, the rolling body peak evaluation matrix and the weight of the rolling body peak evaluation matrix are fuzzily synthesized to obtain a rolling body fuzzy comprehensive evaluation.

[0034] Optionally, if the fuzzy comprehensive evaluation is greater than a fuzzy comprehensive evaluation health threshold, determining that the bearing to be evaluated has a fault includes:

[0035] If the inner ring fuzzy comprehensive evaluation is greater than the inner ring fuzzy comprehensive evaluation health threshold, it is determined that the inner ring fault occurs in the bearing to be evaluated;

[0036] If the outer ring fuzzy comprehensive evaluation is greater than the outer ring fuzzy comprehensive evaluation health threshold, it is determined that the outer ring fault of the bearing to be evaluated occurs;

[0037] If the cage fuzzy comprehensive evaluation is greater than the cage fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a cage failure;

[0038] If the rolling element fuzzy comprehensive evaluation is greater than the rolling element fuzzy comprehensive evaluation health threshold, it is determined that a rolling element failure occurs in the bearing to be evaluated.

[0039] In a second aspect, the present application provides a device for evaluating a bearing fault spectrum, the device comprising:

[0040] A feature extraction module is used to extract features from the frequency spectrum of the bearing to be evaluated, and obtain the characteristic frequency and amplitude of the bearing to be evaluated;

[0041] A determination module, used to determine the theoretical characteristic frequency of the bearing to be evaluated and the neighborhood of the theoretical characteristic frequency, wherein the neighborhood is an adjacent area of ​​the theoretical characteristic frequency;

[0042] A construction module, for constructing a distance evaluation matrix using a Gaussian membership function and the frequencies of the characteristic frequency points of the neighborhood, and for constructing a peak evaluation matrix using a Gaussian membership function and the amplitudes of the characteristic frequency points of the neighborhood;

[0043] The determination module is further used to respectively determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix by using a coefficient of variation method;

[0044] A fuzzy synthesis module, used for fuzzy synthesis of the distance evaluation matrix and the weight of the distance evaluation matrix, the peak evaluation matrix and the weight of the peak evaluation matrix, to obtain a fuzzy comprehensive evaluation;

[0045] The evaluation module is used to determine that the bearing to be evaluated has a fault if the fuzzy comprehensive evaluation is greater than a fuzzy comprehensive evaluation health threshold.

[0046] Optionally, the types of failures of the bearing to be evaluated include inner ring failure, outer ring failure, cage failure and rolling element failure;

[0047] The determination module is further used to determine the inner ring theoretical characteristic frequency of the inner ring fault of the bearing to be evaluated and the neighborhood of the inner ring theoretical characteristic frequency;

[0048] Determine the outer ring theoretical characteristic frequency of the bearing to be evaluated where an outer ring failure occurs and the neighborhood of the outer ring theoretical characteristic frequency;

[0049] Determine the cage theoretical characteristic frequency of the bearing to be evaluated that has a cage failure and the neighborhood of the cage theoretical characteristic frequency;

[0050] Determine the rolling element theoretical characteristic frequency of the bearing to be evaluated when a rolling element failure occurs and the neighborhood of the rolling element theoretical characteristic frequency.

[0051] In a third aspect, an embodiment of the present application provides an electronic device, the electronic device comprising:

[0052] A memory for storing one or more programs;

[0053] Processor; when the one or more programs are executed by the processor, the method for evaluating the bearing fault spectrum described in any one of the first aspects above is implemented.

[0054] In a fourth aspect, an embodiment of the present application provides a computer storage medium, wherein a program is stored in the computer storage medium. When the program is executed by a processor, the method for evaluating a bearing fault spectrum as described in any one of the first aspects above is implemented.

[0055] The above technical solution has the following beneficial effects:

[0056] The embodiment of the present application provides a method, device, equipment and medium for evaluating the spectrum of a bearing fault. When executing the method, the spectrum of the bearing to be evaluated is first feature extracted to obtain the characteristic frequency and amplitude of the bearing to be evaluated; then the theoretical characteristic frequency of the bearing to be evaluated and the neighborhood of the theoretical characteristic frequency are determined; wherein the neighborhood is the adjacent area of ​​the theoretical characteristic frequency; and the Gaussian membership function and the frequency of the characteristic frequency points of the neighborhood are used to construct a distance evaluation matrix, and the Gaussian membership function and the amplitude of the characteristic frequency points of the neighborhood are used to construct a peak evaluation matrix; then the coefficient of variation method is used to determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix respectively; then the distance evaluation matrix and the weight of the distance evaluation matrix, the peak evaluation matrix and the weight of the peak evaluation matrix are fuzzy synthesized to obtain a fuzzy comprehensive evaluation; finally, when the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a fault. In the above manner, the present application takes into account the difference between the characteristic frequency of the theoretical fault of the bearing and the spectrum actually detected, and the evaluation algorithm allows the actual characteristic spectrum to fall within the field of the theoretical spectrum, thereby improving the accuracy of evaluating bearing faults. Moreover, only the neighborhood analysis of theoretical fault characteristic frequency is selected, which is equivalent to data filtering, screening out invalid information and having strong anti-interference ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in this embodiment or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 A flow chart of a method for evaluating a bearing fault spectrum provided in an embodiment of the present application;

[0059] Figure 2 An exemplary schematic diagram of a bearing structure provided in an embodiment of the present application;

[0060] Figure 3 An example schematic diagram of a Gaussian membership function evaluation frequency value provided in an embodiment of the present application;

[0061] Figure 4 An example schematic diagram of the Gaussian membership function evaluation amplitude provided in the embodiment of the present application;

[0062] Figure 5 A schematic diagram of the structure of a bearing fault spectrum evaluation device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] The technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0064] In order to facilitate further understanding of the technical solution provided by this application, the background technology involved in this application will be explained below.

[0065] For motors, bearings are moving parts that are prone to wear, and bearing failures account for about 35% of all motor failures. Monitoring the health of motor bearings helps manage equipment health and increase motor life. The general installation position of bearings is inaccessible and non-visible, and different types of faults can be detected by detecting bearing vibration and winding harmonics.

[0066] Bearing faults can be divided into outer ring faults, inner ring faults, rolling element faults, and cage faults. The basic process of fault diagnosis based on vibration signal spectrum analysis is: signal acquisition → preprocessing → spectrum transformation (feature extraction) → spectrum recognition (feature selection) → diagnosis results. The process of comparing the "similarity" of the actual spectrum with the theoretical spectrum by machine is called feature selection or spectrum recognition.

[0067] The shortcomings of identifying motor bearing fault spectrum are:

[0068] (1) Complex algorithms. Most spectrum algorithms remain at the theoretical research stage and require large amounts of computation. The computing power of MCUs in engineering projects is limited, and most algorithms are not very practical.

[0069] (2) Bearing faults have abundant characteristic signals. For example, the frequency multiples of the vibration acceleration spectrum and the stator coil current spectrum may overlap. The spectra of faults in different parts may also overlap. In this case, the feature selection algorithm may have the problem of inaccurate spectrum identification, which leads to a low accuracy rate in evaluating bearing faults.

[0070] In order to overcome the above technical problems, the embodiment of the present application provides a method for evaluating the spectrum of a bearing fault, firstly extracting the characteristics of the spectrum of the bearing to be evaluated to obtain the characteristic frequency and amplitude of the bearing to be evaluated; then determining the theoretical characteristic frequency of the bearing to be evaluated and the neighborhood of the theoretical characteristic frequency; wherein the neighborhood is the adjacent area of ​​the theoretical characteristic frequency; and constructing a distance evaluation matrix using the Gaussian membership function and the frequency of the characteristic frequency points of the neighborhood, and constructing a peak evaluation matrix using the Gaussian membership function and the amplitude of the characteristic frequency points of the neighborhood; then using the coefficient of variation method to determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix respectively; then fuzzy synthesizing the distance evaluation matrix and the weight of the distance evaluation matrix, the peak evaluation matrix and the weight of the peak evaluation matrix to obtain a fuzzy comprehensive evaluation; finally, when the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a fault. In the above manner, the present application takes into account the difference between the characteristic frequency of the theoretical fault of the bearing and the spectrum actually detected, and the evaluation algorithm allows the actual characteristic spectrum to fall within the field of the theoretical spectrum, thereby improving the accuracy of evaluating bearing faults. Moreover, only the neighborhood analysis of theoretical fault characteristic frequency is selected, which is equivalent to data filtering, screening out invalid information and having strong anti-interference ability.

[0071] See also Figure 1 , Figure 1 A flow chart of a method for evaluating a bearing fault spectrum provided in an embodiment of the present application.

[0072] The method may include:

[0073] Step S101: extracting features from the frequency spectrum of the bearing to be evaluated to obtain the characteristic frequency and amplitude of the bearing to be evaluated.

[0074] Specifically, firstly, the vibration signal of the bearing to be evaluated is subjected to spectrum analysis to obtain the spectrum of the bearing to be evaluated, and then the feature extraction is performed on the spectrum of the bearing to be evaluated to obtain the feature frequency and amplitude of the bearing to be evaluated.

[0075] In specific implementation, the vibration signal can be subjected to autocorrelation analysis, and the analysis result can be subjected to fast Fourier transform (FFT). It can be understood that the autocorrelation analysis can filter out noise, and the autocorrelation spectrum can clearly reflect the frequencies contained in the vibration signal.

[0076] The spectrum of the time domain signal includes the FFT and envelope spectrum of the vibration signal. These two spectra can clearly perceive the impact signal of the bearing in the fault state.

[0077] Step S102: determining the theoretical characteristic frequency of the bearing failure to be evaluated and the neighborhood of the theoretical characteristic frequency, wherein the neighborhood is an adjacent area of ​​the theoretical characteristic frequency.

[0078] In this embodiment, after the spectrum feature is extracted, it is necessary to identify the specific spectrum, and this process is also called feature selection. The machine determines whether a fault has occurred and the type of fault. In theory, as long as the characteristic frequency appears, it can be determined that a fault has occurred. However, due to problems such as lubrication, process equipment, structural deformation, gaps, etc., the actual fault characteristic frequency will appear in the neighborhood of the theoretical characteristic frequency. In addition, when multiple faults occur at the same time, the frequency of secondary phase coupling may also appear. Simple mathematical comparison cannot accurately make a high-precision fault diagnosis.

[0079] In the above case, it is necessary to use fuzzy theory to mathematically quantify all frequency points in the neighborhood of the theoretical characteristic frequency. Apply the fuzzy relationship synthesis principle to quantify some factors with unclear boundaries and difficult to quantify.

[0080] Therefore, the present application further determines the theoretical characteristic frequency of the bearing failure to be evaluated and the neighborhood of the theoretical characteristic frequency, where the neighborhood is the adjacent area of ​​the theoretical characteristic frequency.

[0081] like Figure 2 The figure is a schematic diagram of an example of a bearing structure provided in an embodiment of the present application.

[0082] Since the main bearing structures of the bearing are: inner ring, outer ring, cage, and rolling element, the bearings in use may have inner ring failure, outer ring failure, cage failure, and rolling element failure, which are mostly manifested as wear, deformation, and fracture. The failure point of the damaged bearing will have mechanical impact with other moving parts. By performing spectrum analysis on the vibration signal of the bearing, the failure of the bearing can be better diagnosed according to the characteristics of the spectrum, and the type of failure can be determined.

[0083] Therefore, in a possible implementation, the types of failures of the bearing to be evaluated include inner ring failure, outer ring failure, cage failure and rolling element failure;

[0084] The step of determining the theoretical characteristic frequency of the bearing failure to be evaluated and the neighborhood of the theoretical characteristic frequency includes:

[0085] Determine the inner ring theoretical characteristic frequency of the inner ring fault of the bearing to be evaluated and the neighborhood of the inner ring theoretical characteristic frequency;

[0086] Determine the outer ring theoretical characteristic frequency of the bearing to be evaluated where an outer ring failure occurs and the neighborhood of the outer ring theoretical characteristic frequency;

[0087] Determine the cage theoretical characteristic frequency of the bearing to be evaluated that has a cage failure and the neighborhood of the cage theoretical characteristic frequency;

[0088] Determine the rolling element theoretical characteristic frequency of the bearing to be evaluated when a rolling element failure occurs and the neighborhood of the rolling element theoretical characteristic frequency.

[0089] Specifically, according to theoretical calculations, four types of fault characteristic frequencies can be obtained. Here, the theoretical characteristic frequencies of the outer ring, inner ring, cage, and rolling element of the bearing are set as F bpfo 、F bpfi 、F c 、F bsf , unit is HZ, and the four characteristic frequencies all fall within the low frequency range.

[0090] Bearing F bpfo 、F bpfi 、F c 、F bsf The fundamental frequency is called the characteristic frequency. When a bearing fails, the vibration signal includes not only the fundamental frequency but also the multiples of the fundamental frequency. The multiple coefficients depend on the bearing type and can be integers or decimals. Here, we take the integer multiple coefficients as an example and set them to N. That is, when the actual vibration spectrum (autocorrelation coefficient spectrum, FFT spectrum, envelope spectrum) signal contains kF bpfo , kF bpfi , kF c , kF bsf One or more of them (k=1, 2, 3...N) indicates that the measured bearing has a corresponding type of fault.

[0091] Step S103: constructing a distance evaluation matrix using the Gaussian membership function and the frequencies of the characteristic frequency points of the neighborhood, and constructing a peak evaluation matrix using the Gaussian membership function and the amplitudes of the characteristic frequency points of the neighborhood.

[0092] Specifically, after determining the theoretical characteristic frequency of the bearing failure to be evaluated and the neighborhood of the theoretical characteristic frequency through step S102, it is necessary to use fuzzy theory to mathematically quantify all frequency points in the neighborhood of the theoretical characteristic frequency. Applying the principle of fuzzy relationship synthesis, some factors with unclear boundaries and difficult to quantify are quantified, including: using the Gaussian membership function and the frequency of the characteristic frequency points in the neighborhood to construct a distance evaluation matrix, and using the Gaussian membership function and the amplitude of the characteristic frequency points in the neighborhood to construct a peak evaluation matrix.

[0093] Determine the frequency value and amplitude of all frequency points in the neighborhood of the bearing to be evaluated, the evaluation factors are peak value and distance, and the evaluation set is "size" and "distance". That is, the closer the frequency point is to the characteristic frequency, the better, and the larger the corresponding amplitude is, the better. In summary, the closer the point is to the characteristic frequency, the larger the corresponding amplitude is, and the more likely it is that a fault will occur.

[0094] Among them, the Gaussian function is defined as:

[0095]

[0096] In the formula, δ refers to the standard deviation of the normal distribution, and u refers to the mean in the normal distribution. δ determines the "fatness" of the Gaussian function, that is, the range of the neighborhood; u is the location parameter of the normal distribution, which describes the central tendency position of the normal distribution. The maximum value of this function is 1.

[0097] Perform membership analysis on the frequency values ​​of the frequency points in the neighborhood of the characteristic frequency, and take the characteristic frequency value (including the fundamental frequency and the frequency multiple) as u, that is, u=kF, where k is a positive integer and F is the characteristic frequency. The larger the δ, the larger the range of the neighborhood, the more frequency points are covered, but the frequency selectivity is not strong. The smaller the δ, the narrower the range of the neighborhood, the stronger the frequency selectivity, but it may miss the "effective" frequency information, resulting in inaccurate fault diagnosis. The value of δ is determined according to the needs of on-site adjustment, but no matter how much δ is, the neighborhood range of the frequency should be [u-3δu+3δ].

[0098] Perform membership analysis on the frequency corresponding amplitude of the frequency point in the neighborhood of the characteristic frequency, and take the largest amplitude in the neighborhood as u. The larger the δ, the larger the range of the neighborhood, the more amplitudes are covered, but the amplitude selectivity is not strong. The smaller the δ, the narrower the range of the neighborhood, and the stronger the amplitude selectivity. The value of δ is also determined according to the needs of on-site adjustment.

[0099] The following analysis is based on open source data from the Internet (intercepting the vibration acceleration signal of a faulty bearing with damaged inner and outer rings used on site), and schematically illustrates the evaluation of Gaussian membership on frequency and amplitude.

[0100] like Figure 3 , which is a schematic diagram of an example of evaluating frequency values ​​using a Gaussian membership function provided in an embodiment of the present application.

[0101] exist Figure 3 In the FFT of the vibration signal, the 5th frequency of the inner ring characteristic frequency and the 10th frequency of the outer ring characteristic frequency are taken for analysis. The results show that the actual frequency-multiplied signal does not always appear at the theoretical value, but generally appears near the theoretical value. For example, the peaks of the 2nd frequency of the inner ring and the 5th and 6th frequency of the outer ring appear in the theoretical neighborhood. The frequency membership degree of the frequency value in the 2nd frequency neighborhood of the inner ring of the bearing to the theoretical frequency-multiplied is f(F,2F bpfi ,2), the frequency membership of the frequency values ​​in the inner circle 5 and 6 times the frequency neighborhood to the theoretical 5 and 6 times the frequency are f(F,5F bpfi ,2),f(F,6F bpfi ,2), F is the frequency value in the neighborhood, and the membership of 5 frequency values ​​is calculated for each frequency neighborhood. In each neighborhood, the main factor prominent type (or weighted average type) fuzzy operator is used to synthesize the comprehensive evaluation factor of the frequency value.

[0102] like Figure 4 , which is a schematic diagram of an example of evaluating amplitude using a Gaussian membership function provided in an embodiment of the present application.

[0103] exist Figure 4 In the FFT of the vibration signal, the 5th and 10th times of the outer circle characteristic frequency are taken to evaluate the "peak". The evaluation neighborhood of the frequency value is [u-3δu+3δ]. The values ​​of u and δ are determined by the frequency amplitude contained in the frequency doubling neighborhood. The scientificity of their values ​​will directly affect the size of the evaluation factor. The amplitude membership of the 5th and 10th times of the outer circle characteristic frequency are f(Mag, 0.545, 0.16) and f(Mag, 0.523, 0.12), respectively, where Mag is the amplitude corresponding to the actual frequency point in the neighborhood. The membership of the amplitude corresponding to 5 frequency points is calculated for each of the above frequency doubling neighborhoods. In each neighborhood, the main factor prominence type (or weighted average type) fuzzy operator is used to synthesize the comprehensive evaluation factor of the amplitude.

[0104] In summary, Gaussian functions are used to construct frequency membership functions and amplitude membership functions to evaluate the sum within the neighborhood.

[0105] The specific calculation process is given below:

[0106] Step 1: Evaluation Matrix Construction

[0107] It should be made clear that there are 4 types of faults. Each fault characteristic frequency has N multiples. Each multiple neighborhood contains n frequency points. The frequencies and amplitudes corresponding to these n points are F and ij , Mag ij , (i=1,2,3...N,j=1,2,3...n), the original data matrix is:

[0108]

[0109] (a) Construction of distance evaluation matrix:

[0110] f(F ij ,iF ch ,δ)

[0111] Among them, the δ value determines the frequency selection range; F ch is the characteristic frequency F bpfo 、F bpfi 、F c 、F bsf One of the following; F ij are the elements in the matrix F.

[0112] Construct the distance evaluation matrix of fault types:

[0113]

[0114] Each row represents the frequency evaluation factor of n points in each frequency neighborhood, and there are N rows in total. ij =f(F ij ,iF ch ,δ), where i=1,2,3...N; j=1,2,3...n.

[0115] (b) Peak evaluation matrix construction:

[0116] f(Mag ij ,Mag ch (i),δ)

[0117] Among them, the δ value determines the amplitude selection range; Mag ch (i) is the maximum value of the frequency corresponding to the amplitude in the neighborhood of the i-fold characteristic frequency point, Mag ij is the element in the matrix M;

[0118] Construct the peak evaluation matrix of fault types:

[0119]

[0120] Each row represents the amplitude evaluation factor of n points in each frequency neighborhood, and there are N rows in total. ij =f(Mag ij ,Mag ch (i), δ), where i=1,2,3...N; j=1,2,3...n, δ is given by Mag ch (i) and Mag ij Joint decision.

[0121] The data for the above membership analysis are all spectrum data of the original vibration signal. After the membership calculation, the elements of the evaluation matrix are normalized, and the element sizes of the evaluation matrix R are all in the interval [0,1].

[0122] Step 2: Weight calculation

[0123] (a) Distance matrix standardization

[0124] Distance Matrix:

[0125]

[0126] Normalized distance matrix:

[0127]

[0128] Among them, F n ' z(ij) are the standardized elements of the “distance” matrix, where i=1,2,3...N; j=1,2,3...n.

[0129] (b) Peak matrix normalization

[0130] Normalized peak matrix:

[0131]

[0132] Among them, Mag' nz(ij) are the normalized elements of the “peak” matrix, where i=1,2,3...N; j=1,2,3...n.

[0133] It should be noted that, since the various indicators in the engineering data have different units and ranges, such as the unit of frequency is hz, and the unit of amplitude is acceleration g, the unstandardized data will cause the calculation results to be biased towards indicators with large values. For example, the low-frequency vibration acceleration is large, and the high-frequency acceleration vibration is small, but the variance of the frequency is large, and the variance of the acceleration is not large. If the frequency and acceleration are considered at the same time, the evaluation of the high-frequency calculation result will be "stretched" by the frequency value, but the vibration impact energy of the bearing is concentrated in the low frequency, which will directly lead to unreasonable evaluation. Data standardization ensures that all indicators are in the same dimension, and the linear proportional transformation method is used to perform elementary transformation on the evaluation matrix to obtain a normalized matrix to solve the homogeneity problem of different indicator values, so that they can be compared and calculated fairly.

[0134] Since the bearing may have inner ring fault, outer ring fault, cage fault, and rolling element fault, therefore, in a possible implementation, a distance evaluation matrix is ​​constructed using the Gaussian membership function and the frequency of the characteristic frequency points of the neighborhood, and a peak evaluation matrix is ​​constructed using the Gaussian membership function and the amplitude of the characteristic frequency points of the neighborhood, and further includes:

[0135] An inner circle distance evaluation matrix is ​​constructed using the Gaussian membership function and the frequencies of the characteristic frequency points in the neighborhood of the inner circle theoretical characteristic frequency, and an inner circle peak evaluation matrix is ​​constructed using the Gaussian membership function and the amplitudes of the characteristic frequency points in the neighborhood of the inner circle theoretical characteristic frequency;

[0136] An outer circle distance evaluation matrix is ​​constructed using the Gaussian membership function and the frequencies of the characteristic frequency points in the neighborhood of the outer circle theoretical characteristic frequency, and an outer circle peak evaluation matrix is ​​constructed using the Gaussian membership function and the amplitudes of the characteristic frequency points in the neighborhood of the outer circle theoretical characteristic frequency;

[0137] Constructing a cage distance evaluation matrix using the Gaussian membership function and the frequencies of the characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the cage, and constructing a cage peak evaluation matrix using the Gaussian membership function and the amplitudes of the characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the cage;

[0138] A rolling element distance evaluation matrix is ​​constructed using the Gaussian membership function and the frequencies of characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the rolling element, and a rolling element peak evaluation matrix is ​​constructed using the Gaussian membership function and the amplitudes of characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the rolling element.

[0139] Step S104: using the coefficient of variation method to respectively determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix.

[0140] In this step, different types of n points corresponding to different multiples have two evaluation factors, namely "distance" and "peak value". When the final comprehensive evaluation factor is obtained by using the synthesis operator, their weights are different.

[0141] Therefore, in this example, the coefficient of variation method is also used to respectively determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix.

[0142] (c) Calculate weights

[0143] Specifically, F' nz (ij) represents F' nz The element in the i-th row and j-th column of nz (ij) indicates Mag' nz The weight of the element in the i-th row and j-th column is determined using the coefficient of variation method.

[0144] The mean and standard deviation of the distance evaluation matrix are:

[0145]

[0146] The mean and standard deviation of the amplitude matrix are:

[0147]

[0148] Among them, i=1,2,3...N, j=1,2,3...n.

[0149] Then the normalized weight of the distance index is:

[0150] wxya i =vF i / ∑v

[0151] Then the normalized weight of the peak index is:

[0152] wxya i =vM i / ∑v

[0153] in, ∑v=(vF i +vM i ), i = 1, 2, 3...N. Each octave has its own distance and peak index weight.

[0154] Since the bearing may have inner ring failure, outer ring failure, cage failure, and rolling element failure, therefore, in a possible implementation, the weight of the distance evaluation matrix and the weight of the peak evaluation matrix are determined by using the coefficient of variation method, including:

[0155] Using the coefficient of variation method, respectively determining the weight of the inner circle distance evaluation matrix and the weight of the inner circle peak value evaluation matrix;

[0156] Using the coefficient of variation method, respectively determining the weight of the outer circle distance evaluation matrix and the weight of the outer circle peak evaluation matrix;

[0157] Using a coefficient of variation method, respectively determining the weight of the cage distance evaluation matrix and the weight of the cage peak evaluation matrix;

[0158] The weight of the rolling element distance evaluation matrix and the weight of the rolling element peak value evaluation matrix are determined respectively by using the coefficient of variation method.

[0159] Step 3: Blur Synthesis

[0160] Step S105: fuzzy synthesizing the distance evaluation matrix and the weight of the distance evaluation matrix, and the peak evaluation matrix and the weight of the peak evaluation matrix to obtain a fuzzy comprehensive evaluation.

[0161] Specifically, first select the "product-sum" fuzzy operator. The matrix of the N-fold fuzzy synthesis of the characteristic frequency of a certain fault type is:

[0162]

[0163] in, (i=1,2,3...N, j=1,2,3...n). j i Fuzzy evaluation of the i-fold frequency of a certain characteristic fault.

[0164] Since the proportion of frequency doublings in each stage of the bearing is different, the frequency doublings in the initial stage are not very obvious; the proportion of frequency doublings in the middle stage will increase significantly; and the characteristic frequency and its frequency doublings in the final stage will be very prominent.

[0165] Since it is impossible to directly obtain the specific frequency multiple occurrence ratio, the same frequency multiple ratio is taken here, so the fuzzy comprehensive evaluation of N frequency multiples of a certain characteristic fault should be: Using fuzzy comprehensive evaluation to evaluate health threshold ref is the threshold value, when J sum >J ref When , this type of failure is considered to have occurred. And J sum The larger it is, the more serious the fault is.

[0166] Since there are four types of faults in faulty bearings, in a possible implementation, the distance evaluation matrix and the weight of the distance evaluation matrix, the peak evaluation matrix and the weight of the peak evaluation matrix are fuzzy synthesized to obtain a fuzzy comprehensive evaluation, including:

[0167] Performing fuzzy synthesis on the inner circle distance evaluation matrix and the weight of the inner circle distance evaluation matrix, the inner circle peak evaluation matrix and the weight of the inner circle peak evaluation matrix to obtain an inner circle fuzzy comprehensive evaluation;

[0168] Performing fuzzy synthesis on the outer circle distance evaluation matrix and the weight of the outer circle distance evaluation matrix, the outer circle peak evaluation matrix and the weight of the outer circle peak evaluation matrix to obtain an outer circle fuzzy comprehensive evaluation;

[0169] Performing fuzzy synthesis on the cage distance evaluation matrix and the weight of the cage distance evaluation matrix, the cage peak evaluation matrix and the weight of the cage peak evaluation matrix to obtain a cage fuzzy comprehensive evaluation;

[0170] The rolling body distance evaluation matrix and the weight of the rolling body distance evaluation matrix, the rolling body peak evaluation matrix and the weight of the rolling body peak evaluation matrix are fuzzily synthesized to obtain a rolling body fuzzy comprehensive evaluation.

[0171] Step S106: If the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated is faulty.

[0172] In this embodiment, when the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a fault, wherein the fuzzy comprehensive evaluation health threshold can be set according to actual conditions and is not limited here.

[0173] It can be seen from the above technical scheme that the present application first extracts the characteristics of the frequency spectrum of the bearing to be evaluated to obtain the characteristic frequency and amplitude of the bearing to be evaluated; then determines the theoretical characteristic frequency of the bearing to be evaluated and the neighborhood of the theoretical characteristic frequency; wherein the neighborhood is the adjacent area of ​​the theoretical characteristic frequency; and uses the Gaussian membership function and the frequency of the characteristic frequency points of the neighborhood to construct a distance evaluation matrix, and uses the Gaussian membership function and the amplitude of the characteristic frequency points of the neighborhood to construct a peak evaluation matrix; then uses the coefficient of variation method to determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix respectively; then fuzzy synthesizes the distance evaluation matrix and the weight of the distance evaluation matrix, the peak evaluation matrix and the weight of the peak evaluation matrix to obtain a fuzzy comprehensive evaluation; finally, when the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a fault. In the above manner, the present application takes into account the difference between the characteristic frequency of the theoretical fault of the bearing and the spectrum actually detected, and the evaluation algorithm allows the actual characteristic spectrum to fall within the field of the theoretical spectrum, thereby improving the accuracy of evaluating bearing faults. Moreover, only the neighborhood analysis of theoretical fault characteristic frequency is selected, which is equivalent to data filtering, screening out invalid information and having strong anti-interference ability.

[0174] Since there are four types of faults in faulty bearings, in a possible implementation, if the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a fault, including:

[0175] If the inner ring fuzzy comprehensive evaluation is greater than the inner ring fuzzy comprehensive evaluation health threshold, it is determined that the inner ring fault occurs in the bearing to be evaluated;

[0176] If the outer ring fuzzy comprehensive evaluation is greater than the outer ring fuzzy comprehensive evaluation health threshold, it is determined that the outer ring fault of the bearing to be evaluated occurs;

[0177] If the cage fuzzy comprehensive evaluation is greater than the cage fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a cage failure;

[0178] If the rolling element fuzzy comprehensive evaluation is greater than the rolling element fuzzy comprehensive evaluation health threshold, it is determined that a rolling element failure occurs in the bearing to be evaluated, and finally 4N fuzzy comprehensive evaluations are obtained through analysis.

[0179] It should be noted that the embodiment of the present application takes into account the difference between the theoretical characteristic frequency of the bearing fault and the actual detected spectrum. The detection algorithm allows the actual characteristic spectrum to fall within the field of the theoretical spectrum, thereby improving the detection accuracy. Moreover, the frequency and amplitude of all points in the neighborhood of the fault characteristic frequency are evaluated, and the base frequency and multiple frequency are comprehensively evaluated in multiple dimensions, so that the information is fully utilized.

[0180] For the above-mentioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should know that the present invention is not limited by the order of the actions described, because according to the present invention, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present invention.

[0181] The above are some specific implementations of the bearing fault spectrum evaluation method provided in the embodiment of the present application. Based on this, the present application also provides a corresponding device. The following will introduce the bearing fault spectrum evaluation device provided in the embodiment of the present application from the perspective of functional modularization.

[0182] See also Figure 5 , is a structural schematic diagram of a bearing fault spectrum evaluation device provided in an embodiment of the present application, the device comprising:

[0183] A feature extraction module 501 is used to extract features from the frequency spectrum of the bearing to be evaluated, and obtain the characteristic frequency and amplitude of the bearing to be evaluated;

[0184] A determination module 502 is used to determine the theoretical characteristic frequency of the bearing to be evaluated and the neighborhood of the theoretical characteristic frequency, where the neighborhood is an adjacent area of ​​the theoretical characteristic frequency;

[0185] A construction module 503 is used to construct a distance evaluation matrix using a Gaussian membership function and the frequencies of the characteristic frequency points of the neighborhood, and to construct a peak evaluation matrix using a Gaussian membership function and the amplitudes of the characteristic frequency points of the neighborhood;

[0186] The determination module 502 is further used to respectively determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix by using a coefficient of variation method;

[0187] A fuzzy synthesis module 504 is used to perform fuzzy synthesis on the distance evaluation matrix and the weight of the distance evaluation matrix, and the peak evaluation matrix and the weight of the peak evaluation matrix to obtain a fuzzy comprehensive evaluation;

[0188] The evaluation module 505 is used to determine that the bearing to be evaluated has a fault if the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold.

[0189] Optionally, the types of failures of the bearing to be evaluated include inner ring failure, outer ring failure, cage failure and rolling element failure;

[0190] The determination module 502 is further used to determine the inner ring theoretical characteristic frequency of the inner ring fault of the bearing to be evaluated and the neighborhood of the inner ring theoretical characteristic frequency;

[0191] Determine the outer ring theoretical characteristic frequency of the bearing to be evaluated where an outer ring failure occurs and the neighborhood of the outer ring theoretical characteristic frequency;

[0192] Determine the cage theoretical characteristic frequency of the bearing to be evaluated that has a cage failure and the neighborhood of the cage theoretical characteristic frequency;

[0193] Determine the rolling element theoretical characteristic frequency of the bearing to be evaluated when a rolling element failure occurs and the neighborhood of the rolling element theoretical characteristic frequency.

[0194] It can be seen from the above technical scheme that the embodiment of the present application first extracts the characteristics of the frequency spectrum of the bearing to be evaluated to obtain the characteristic frequency and amplitude of the bearing to be evaluated; then determines the theoretical characteristic frequency of the bearing to be evaluated and the neighborhood of the theoretical characteristic frequency; wherein the neighborhood is the adjacent area of ​​the theoretical characteristic frequency; and uses the Gaussian membership function and the frequency of the characteristic frequency points of the neighborhood to construct a distance evaluation matrix, and uses the Gaussian membership function and the amplitude of the characteristic frequency points of the neighborhood to construct a peak evaluation matrix; then uses the coefficient of variation method to determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix respectively; then fuzzy synthesizes the distance evaluation matrix and the weight of the distance evaluation matrix, and the peak evaluation matrix and the weight of the peak evaluation matrix to obtain a fuzzy comprehensive evaluation; finally, when the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a fault. In the above manner, the present application takes into account the difference between the characteristic frequency of the theoretical fault of the bearing and the spectrum actually detected, and the evaluation algorithm allows the actual characteristic spectrum to fall within the field of the theoretical spectrum, thereby improving the accuracy of evaluating bearing faults. Moreover, only the neighborhood analysis of theoretical fault characteristic frequency is selected, which is equivalent to data filtering, screening out invalid information and having strong anti-interference ability.

[0195] An embodiment of the present application further provides an electronic device, comprising: a memory for storing one or more programs;

[0196] Processor; when the one or more programs are executed by the processor, the bearing fault spectrum evaluation method described in the above embodiment is implemented.

[0197] The embodiment of the present application further provides a computer storage medium, in which a program is stored. When the program is executed by a processor, the method for evaluating the bearing fault spectrum described in the above embodiment is implemented.

[0198] The "first" and "second" in the names such as "first" and "second" (if any) mentioned in the embodiments of the present application are only used as name identifiers and do not represent the first or second in order.

[0199] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0200] Those skilled in the art can understand that the flowchart shown in the figure is only an example in which the embodiments of the present application can be implemented, and the scope of application of the embodiments of the present application is not limited by any aspect of the flowchart.

[0201] In the several embodiments provided in the present application, it should be understood that the disclosed methods, devices and equipment can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.

[0202] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, the functional units in the various embodiments of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0203] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0204] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for evaluating a bearing fault spectrum, characterized in that: The method comprises: Extracting features from the frequency spectrum of the bearing to be evaluated to obtain the characteristic frequency and amplitude of the bearing to be evaluated; Determine the theoretical characteristic frequency of the bearing to be evaluated and the neighborhood of the theoretical characteristic frequency, wherein the neighborhood is an adjacent area of ​​the theoretical characteristic frequency; Constructing a distance evaluation matrix using a Gaussian membership function and the frequencies of the characteristic frequency points of the neighborhood, and constructing a peak evaluation matrix using a Gaussian membership function and the amplitudes of the characteristic frequency points of the neighborhood; Determining the weight of the distance evaluation matrix and the weight of the peak evaluation matrix respectively by using a coefficient of variation method; Performing fuzzy synthesis on the distance evaluation matrix and the weight of the distance evaluation matrix, and the peak evaluation matrix and the weight of the peak evaluation matrix to obtain a fuzzy comprehensive evaluation; If the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a fault.

2. The method according to claim 1, characterized in that The types of bearing failures to be evaluated include inner ring failure, outer ring failure, cage failure and rolling element failure; The step of determining the theoretical characteristic frequency of the bearing failure to be evaluated and the neighborhood of the theoretical characteristic frequency includes: Determine the inner ring theoretical characteristic frequency of the inner ring fault of the bearing to be evaluated and the neighborhood of the inner ring theoretical characteristic frequency; Determine the outer ring theoretical characteristic frequency of the bearing to be evaluated where an outer ring failure occurs and the neighborhood of the outer ring theoretical characteristic frequency; Determine the cage theoretical characteristic frequency of the bearing to be evaluated that has a cage failure and the neighborhood of the cage theoretical characteristic frequency; Determine the rolling element theoretical characteristic frequency of the bearing to be evaluated when a rolling element failure occurs and the neighborhood of the rolling element theoretical characteristic frequency.

3. The method according to claim 2, characterized in that The method of constructing a distance evaluation matrix using the Gaussian membership function and the frequencies of the characteristic frequency points of the neighborhood, and constructing a peak evaluation matrix using the Gaussian membership function and the amplitudes of the characteristic frequency points of the neighborhood, includes: An inner circle distance evaluation matrix is ​​constructed using the Gaussian membership function and the frequencies of the characteristic frequency points in the neighborhood of the inner circle theoretical characteristic frequency, and an inner circle peak evaluation matrix is ​​constructed using the Gaussian membership function and the amplitudes of the characteristic frequency points in the neighborhood of the inner circle theoretical characteristic frequency; An outer circle distance evaluation matrix is ​​constructed using the Gaussian membership function and the frequencies of the characteristic frequency points in the neighborhood of the outer circle theoretical characteristic frequency, and an outer circle peak evaluation matrix is ​​constructed using the Gaussian membership function and the amplitudes of the characteristic frequency points in the neighborhood of the outer circle theoretical characteristic frequency; Constructing a cage distance evaluation matrix using the Gaussian membership function and the frequencies of the characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the cage, and constructing a cage peak evaluation matrix using the Gaussian membership function and the amplitudes of the characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the cage; A rolling element distance evaluation matrix is ​​constructed using the Gaussian membership function and the frequencies of characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the rolling element, and a rolling element peak evaluation matrix is ​​constructed using the Gaussian membership function and the amplitudes of characteristic frequency points in the neighborhood of the theoretical characteristic frequency of the rolling element.

4. The method according to claim 3, characterized in that The method of using the coefficient of variation method to respectively determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix includes: Using the coefficient of variation method, respectively determining the weight of the inner circle distance evaluation matrix and the weight of the inner circle peak value evaluation matrix; Using the coefficient of variation method, respectively determining the weight of the outer circle distance evaluation matrix and the weight of the outer circle peak evaluation matrix; Using a coefficient of variation method, respectively determining the weight of the cage distance evaluation matrix and the weight of the cage peak evaluation matrix; The weight of the rolling element distance evaluation matrix and the weight of the rolling element peak value evaluation matrix are determined respectively by using the coefficient of variation method.

5. The method according to claim 4, characterized in that The fuzzy synthesis of the distance evaluation matrix and the weight of the distance evaluation matrix, the peak evaluation matrix and the weight of the peak evaluation matrix to obtain a fuzzy comprehensive evaluation includes: Performing fuzzy synthesis on the inner circle distance evaluation matrix and the weight of the inner circle distance evaluation matrix, the inner circle peak evaluation matrix and the weight of the inner circle peak evaluation matrix to obtain an inner circle fuzzy comprehensive evaluation; Performing fuzzy synthesis on the outer circle distance evaluation matrix and the weight of the outer circle distance evaluation matrix, the outer circle peak evaluation matrix and the weight of the outer circle peak evaluation matrix to obtain an outer circle fuzzy comprehensive evaluation; Performing fuzzy synthesis on the cage distance evaluation matrix and the weight of the cage distance evaluation matrix, the cage peak evaluation matrix and the weight of the cage peak evaluation matrix to obtain a cage fuzzy comprehensive evaluation; The rolling body distance evaluation matrix and the weight of the rolling body distance evaluation matrix, the rolling body peak evaluation matrix and the weight of the rolling body peak evaluation matrix are fuzzily synthesized to obtain a rolling body fuzzy comprehensive evaluation.

6. The method according to claim 5, characterized in that If the fuzzy comprehensive evaluation is greater than the fuzzy comprehensive evaluation health threshold, determining that the bearing to be evaluated has a fault includes: If the inner ring fuzzy comprehensive evaluation is greater than the inner ring fuzzy comprehensive evaluation health threshold, it is determined that the inner ring fault occurs in the bearing to be evaluated; If the outer ring fuzzy comprehensive evaluation is greater than the outer ring fuzzy comprehensive evaluation health threshold, it is determined that the outer ring fault of the bearing to be evaluated occurs; If the cage fuzzy comprehensive evaluation is greater than the cage fuzzy comprehensive evaluation health threshold, it is determined that the bearing to be evaluated has a cage failure; If the rolling element fuzzy comprehensive evaluation is greater than the rolling element fuzzy comprehensive evaluation health threshold, it is determined that a rolling element failure occurs in the bearing to be evaluated.

7. A device for evaluating a bearing fault spectrum, characterized in that: The device comprises: A feature extraction module is used to extract features from the frequency spectrum of the bearing to be evaluated, and obtain the characteristic frequency and amplitude of the bearing to be evaluated; A determination module, used to determine the theoretical characteristic frequency of the bearing to be evaluated and the neighborhood of the theoretical characteristic frequency, wherein the neighborhood is an adjacent area of ​​the theoretical characteristic frequency; A construction module, for constructing a distance evaluation matrix using a Gaussian membership function and the frequencies of the characteristic frequency points of the neighborhood, and for constructing a peak evaluation matrix using a Gaussian membership function and the amplitudes of the characteristic frequency points of the neighborhood; The determination module is further used to respectively determine the weight of the distance evaluation matrix and the weight of the peak evaluation matrix by using a coefficient of variation method; A fuzzy synthesis module, used for fuzzy synthesis of the distance evaluation matrix and the weight of the distance evaluation matrix, the peak evaluation matrix and the weight of the peak evaluation matrix, to obtain a fuzzy comprehensive evaluation; The evaluation module is used to determine that the bearing to be evaluated has a fault if the fuzzy comprehensive evaluation is greater than a fuzzy comprehensive evaluation health threshold.

8. The device according to claim 7, characterized in that The types of bearing failures to be evaluated include inner ring failure, outer ring failure, cage failure and rolling element failure; The determination module is further used to determine the inner ring theoretical characteristic frequency of the inner ring fault of the bearing to be evaluated and the neighborhood of the inner ring theoretical characteristic frequency; Determine the outer ring theoretical characteristic frequency of the bearing to be evaluated where an outer ring failure occurs and the neighborhood of the outer ring theoretical characteristic frequency; Determine the cage theoretical characteristic frequency of the bearing to be evaluated that has a cage failure and the neighborhood of the cage theoretical characteristic frequency; Determine the rolling element theoretical characteristic frequency of the bearing to be evaluated when a rolling element failure occurs and the neighborhood of the rolling element theoretical characteristic frequency.

9. An electronic device, characterized in that: The electronic device comprises: A memory for storing one or more programs; Processor; when the one or more programs are executed by the processor, the method for evaluating the bearing fault spectrum as described in any one of claims 1-6 is implemented.

10. A computer storage medium, characterized in that: The computer storage medium stores a program, and when the program is executed by the processor, the method for evaluating the bearing fault spectrum according to any one of claims 1 to 6 is implemented.

Images