Bearing fault diagnosis method, device and equipment and storage medium

Abstract

The embodiment of the invention provides a bearing fault diagnosis method and device, equipment and a storage medium. The method comprises the following steps: acquiring a vibration signal of a bearing to be detected, a time domain feature of the vibration signal and a frequency domain feature of the vibration signal; splicing the time domain feature and the frequency domain feature to obtain a spliced feature of the vibration signal; inputting the spliced features into a diagnosis model to obtain reconstructed features; according to a reconstruction error between the reconstruction feature and the splicing feature and an error threshold value, whether the bearing to be detected has a fault is judged, and the error threshold value is determined according to a reconstruction error of a second sample set in the diagnosis model; the second sample set comprises a vibration signal of the second bearing in a healthy state and a vibration signal of the third bearing in a fault state. According to the embodiment of the invention, the problem that a fixed threshold cannot adapt to diversified data distribution in a traditional method is effectively solved, and the detection performance in different scenes is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a bearing fault diagnosis method, device, equipment and storage medium. Background Art

[0002] Bearings are vital components in industrial machinery and equipment. Their operating status directly affects the safety and reliability of the equipment. Once a bearing fails, it may cause equipment downtime, production interruption, and even cause serious economic losses and safety accidents. Therefore, timely and accurate detection of the health status of bearings and early warning of potential failures are the key to ensuring the safe operation of equipment. However, traditional bearing anomaly detection methods usually rely on fixed thresholds set by humans. This method requires rich domain knowledge and faces many challenges in practical applications.

[0003] Under different working conditions, the characteristics of bearing vibration signals vary greatly. If the threshold is set too high, missed detection may occur. If the threshold is set too low, a high false alarm rate will occur, seriously affecting the detection effect. Therefore, how to improve the accuracy of bearing fault detection has become an important research topic. Summary of the invention

[0004] The embodiments of the present invention provide a bearing fault diagnosis method, device, equipment and storage medium, which can dynamically determine an error threshold, thereby improving the accuracy of bearing fault diagnosis.

[0005] In a first aspect, an embodiment of the present invention provides a bearing fault diagnosis method, the method comprising:

[0006] Acquire a vibration signal of the bearing to be tested, a time domain feature of the vibration signal, and a frequency domain feature of the vibration signal;

[0007] Splicing the time domain features and the frequency domain features to obtain a splicing feature of the vibration signal;

[0008] Inputting the spliced ​​features into a diagnostic model to obtain a reconstructed feature, wherein the diagnostic model is obtained by training a first sample set, the first sample set includes training samples, and the training samples are vibration signals of the first bearing in a healthy state;

[0009] Based on the reconstruction error between the reconstructed feature and the splicing feature and the error threshold, it is determined whether the bearing to be tested is faulty, wherein the error threshold is determined based on the reconstruction error generated after the second sample set is input into the diagnostic model, and the second sample set includes the vibration signal of the second bearing in a healthy state and the vibration signal of the third bearing in a faulty state.

[0010] Furthermore, the error threshold is obtained by the following steps:

[0011] Obtaining a reconstruction error corresponding to each vibration signal in the second sample set according to the splicing feature corresponding to each vibration signal in the second sample set and the diagnostic model;

[0012] classifying the reconstruction error corresponding to each vibration signal in the second sample set to obtain a normal category;

[0013] The reconstruction error corresponding to the upper quartile in the normal category is used as the error threshold.

[0014] Furthermore, the diagnostic model is an autoencoder, and the diagnostic model is an autoencoder. The training process of the diagnostic model is as follows:

[0015] Using the vibration signal of the first bearing within a preset historical time period as the training sample;

[0016] Inputting the concatenated features corresponding to the training samples into the autoencoder to obtain the reconstructed features corresponding to the training samples;

[0017] By using a stochastic gradient descent algorithm, with the goal of minimizing the reconstruction error between the splicing features corresponding to the training samples and the reconstructed features corresponding to the training samples, the training parameters of the autoencoder are updated until the training end condition is reached, thereby obtaining the diagnostic model.

[0018] Furthermore, the vibration signal includes a plurality of vibration data, and the time domain characteristics of the vibration signal are obtained by the following steps:

[0019] According to the number of rotations of the bearing to be tested, the plurality of vibration data are divided into a plurality of vibration subsequences;

[0020] According to the time domain statistical characteristics of each vibration subsequence, the time domain characteristics corresponding to the vibration signal are obtained, wherein the time domain statistical characteristics include at least one of a mean, a variance, a skewness, a kurtosis, a coefficient of variation, a maximum amplitude, a minimum amplitude, an upper quartile, and a lower quartile.

[0021] Further, the vibration signal includes a plurality of vibration data, and the frequency domain characteristics of the vibration signal are obtained by the following steps:

[0022] According to the number of rotations of the bearing to be tested, the plurality of vibration data are divided into a plurality of vibration subsequences;

[0023] After converting each vibration subsequence into the frequency domain, the power spectrum density corresponding to each vibration subsequence is calculated;

[0024] According to the frequency domain statistical characteristics of each power spectrum density, the frequency domain characteristics corresponding to the vibration signal are obtained, wherein the frequency domain statistical characteristics include at least one of average frequency, frequency variance, root mean square frequency and center frequency.

[0025] Furthermore, the step of splicing the time domain feature and the frequency domain feature to obtain the splicing feature of the vibration signal includes:

[0026] Compressing the time domain features and the frequency domain features into one-dimensional vectors respectively;

[0027] The compressed time domain features and the compressed frequency domain features are spliced ​​to obtain the spliced ​​features.

[0028] Furthermore, before the step of inputting the splicing features into the diagnosis model to obtain the reconstruction features, the method further comprises:

[0029] The splicing features are normalized.

[0030] In a second aspect, an embodiment of the present invention provides a bearing fault diagnosis device, the device comprising:

[0031] An acquisition module, used to acquire the vibration signal of the bearing to be tested, the time domain characteristics of the vibration signal, and the frequency domain characteristics of the vibration signal;

[0032] A splicing module, used for splicing the time domain features and the frequency domain features to obtain a splicing feature of the vibration signal;

[0033] A reconstruction module, used for inputting the splicing features into a diagnostic model to obtain a reconstructed feature, wherein the diagnostic model is obtained by training a first sample set, the first sample set includes training samples, and the training samples are vibration signals of the first bearing in a healthy state;

[0034] A diagnostic module is used to determine whether the bearing to be tested has a fault based on a reconstruction error between the reconstructed feature and the spliced ​​feature and an error threshold, wherein the error threshold is determined based on a reconstruction error generated after a second sample set is input into the diagnostic model, and the second sample set includes a vibration signal of the second bearing in a healthy state and a vibration signal of the third bearing in a faulty state.

[0035] In a third aspect, an embodiment of the present invention provides a computer device, the device comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0036] Memory, used to store computer programs;

[0037] The processor is used to implement the steps of a bearing fault diagnosis method provided in the first aspect when executing the program stored in the memory.

[0038] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps of a bearing fault diagnosis method provided in the first aspect are implemented.

[0039] The embodiments of the present invention provide a bearing fault diagnosis method, device, equipment and storage medium, which splice the time domain characteristics and frequency domain characteristics of the vibration signal of the bearing to be tested to obtain spliced ​​characteristics; then input the spliced ​​characteristics into a diagnosis model to obtain a reconstructed feature; finally, determine whether the bearing to be tested has a fault based on the reconstruction error and the error threshold.

[0040] In the embodiment of the present invention, the diagnostic model is trained by the vibration signal of the first bearing in a healthy state. Therefore, the vibration signal in an abnormal state will produce a large reconstruction error in the diagnostic model, so that the normal state and the fault state can be distinguished according to the reconstruction error; in addition, the error threshold is determined according to the reconstruction error corresponding to the vibration signal in a healthy state and the vibration signal in a fault state. The error threshold can be dynamically adjusted according to the characteristics of the vibration signal at different positions, so that the error threshold is more accurate, ensuring the sensitivity and accuracy of bearing fault detection. The embodiment of the present invention effectively solves the problem that the fixed threshold in the traditional method cannot adapt to the distribution of diversified data, and improves the detection performance in different scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0042] Figure 1 A flow chart of a bearing fault diagnosis method is provided for an embodiment of the present invention;

[0043] Figure 2 A structural schematic diagram of a bearing fault diagnosis device is provided for an embodiment of the present invention;

[0044] Figure 3 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0046] In order to make the description of the disclosed content more detailed and complete, the following is an illustrative description of the implementation mode and specific examples of the present invention; however, this is not the only form of implementing or applying the specific embodiments of the present invention. The implementation mode covers the features of multiple specific embodiments and the method steps and their sequence for constructing and operating these specific embodiments. However, other specific embodiments can also be used to achieve the same or equal functions and step sequences. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0047] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein.

[0048] In the description of the embodiments of the present invention, unless otherwise specified, “ / ” means or, for example, A / B can mean A or B; “and / or” in the text is merely a description of the association relationship of associated objects, indicating that three relationships may exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, “multiple” refers to two or more than two, and other quantifiers are similar. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention, and the embodiments of the present application and the features in the embodiments may be combined with each other without conflict.

[0049] An embodiment of the present invention provides a bearing fault diagnosis method, which can be applied to various scenarios of bearing fault detection, such as industrial machinery and equipment, wind turbines, medical equipment, railway transportation, and automobile manufacturing. When bearing fault diagnosis is required, the client sends a bearing fault diagnosis signal, and after the server receives the bearing fault diagnosis signal, it first obtains the vibration signal of the bearing to be tested, the time domain characteristics of the vibration signal, and the frequency domain characteristics of the vibration signal; splices the time domain characteristics and the frequency domain characteristics to obtain the spliced ​​characteristics of the vibration signal; then inputs the spliced ​​characteristics into the diagnosis model to obtain the reconstructed characteristics; finally, based on the reconstruction error between the reconstructed characteristics and the spliced ​​characteristics and the size of the error threshold, it is judged whether the bearing to be tested has a fault.

[0050] The server can be implemented by an independent server or a server cluster composed of multiple servers. The client can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The client and the server can be connected via Bluetooth, USB (Universal Serial Bus) or other communication connection methods, which is not limited in the embodiments of the present invention.

[0051] Figure 1 A flow chart of a bearing fault diagnosis method is provided for an embodiment of the present invention, such as Figure 1 As shown, the method includes:

[0052] S110, obtaining a vibration signal of the bearing to be tested, a time domain feature of the vibration signal, and a frequency domain feature of the vibration signal;

[0053] First, obtain the vibration signal of the bearing to be tested, which is the bearing that needs to be tested for fault detection; in the process of collecting vibration signals, the embodiment of the present invention pre-installs sensors at multiple detection points of the bearing to be tested, and collects the vibration signals generated by the bearing during operation through the sensors. Specifically, taking one of the detection points as an example, during the process of the bearing to be tested rotating one circle, m vibration data are collected according to the sampling frequency corresponding to the detection point. If the bearing to be tested rotates n circles, the detection point can collect n*m vibration data. Therefore, the vibration signal corresponding to the detection point includes m×n vibration data. The vibration signal corresponding to each detection point is collected in the same way.

[0054] Among them, m and n are both positive integers, and the specific values ​​can be determined according to actual conditions, and the embodiment of the present invention does not make specific limitations on this; the sampling frequency can be 128KHz or 256KHz, which can be determined according to actual conditions, and the embodiment of the present invention does not make specific limitations on this.

[0055] After collecting the vibration signal corresponding to each detection point of the bearing to be tested, the time domain characteristics and frequency domain characteristics of each vibration signal are calculated. For example, when calculating the time domain characteristics of the vibration signal, the mean, variance, etc. can be calculated, and when calculating the frequency domain characteristics of the vibration signal, the average frequency, average variance, etc. can be calculated. The specific characteristics can be determined according to actual conditions, and the embodiments of the present invention do not make specific limitations on this.

[0056] S120, splicing the time domain feature and the frequency domain feature to obtain a spliced ​​feature of the vibration signal;

[0057] Then, the time domain features and frequency domain features are spliced ​​to obtain the spliced ​​features of the vibration signal. Feature splicing is an operation that combines multiple features into a new feature vector. Feature splicing can integrate features from different data sources or that have been processed differently, providing richer information for subsequent diagnostic model training.

[0058] S130, inputting the spliced ​​features into a diagnostic model to obtain a reconstructed feature, wherein the diagnostic model is obtained by training a first sample set, the first sample set includes training samples, and the training samples are vibration signals of the first bearing in a healthy state;

[0059] Then, the spliced ​​features are input into the diagnostic model, and the diagnostic model outputs the reconstructed features. In the embodiment of the present invention, the diagnostic model is trained by the training samples in the first sample set, and the training samples only refer to the vibration signal of the first bearing in a healthy state, and do not include the vibration signal in a faulty state. Since the diagnostic model has only learned the vibration signal in a healthy state, when the vibration signal in a faulty state is input into the diagnostic model, a large reconstruction error will be generated, while the reconstruction error generated when the vibration signal in a healthy state is input into the diagnostic model is small. Therefore, it is possible to determine whether the bearing is faulty based on the size of the reconstruction error.

[0060] S140, judging whether the bearing to be tested is faulty based on a reconstruction error between the reconstructed feature and the spliced ​​feature and an error threshold, wherein the error threshold is determined based on a reconstruction error generated after a second sample set is input into the diagnostic model, and the second sample set includes a vibration signal of the second bearing in a healthy state and a vibration signal of the third bearing in a faulty state.

[0061] Finally, the reconstruction error between the reconstructed features and the spliced ​​features is calculated, the size of the reconstruction error and the error threshold is compared, and whether the bearing to be tested is faulty is determined based on the comparison result. In the embodiment of the present invention, the error threshold is determined based on the reconstruction error generated by the second sample set input into the diagnostic model, and the second sample set includes both the vibration signal of the second bearing in a healthy state and the vibration signal of the third bearing in a faulty state, so that the error threshold can be dynamically adjusted according to the characteristics of the vibration signals at different positions, so that the error threshold is more accurate, ensuring the sensitivity and accuracy of bearing fault detection.

[0062] The reconstruction error calculation formula is: x represents the splicing feature, represents the reconstruction feature, ||·|| 2 represents the L2 norm.

[0063] In practical applications, when the reconstruction error is greater than the error threshold, the bearing under test is judged to be faulty; otherwise, it is judged to be normal. Through the above adaptive threshold method, the threshold setting can be dynamically adjusted according to the characteristics of vibration signals at different positions to ensure the sensitivity and accuracy of anomaly detection. This method effectively solves the problem that traditional fixed thresholds cannot adapt to diversified data distribution and improves the detection performance of the model in different scenarios.

[0064] An embodiment of the present invention provides a bearing fault diagnosis method, which splices the time domain characteristics and frequency domain characteristics of the vibration signal of the bearing to be tested to obtain a spliced ​​feature; then inputs the spliced ​​feature into a diagnosis model to obtain a reconstructed feature; finally, determines whether the bearing to be tested has a fault based on the reconstruction error and the error threshold.

[0065] In the embodiment of the present invention, the diagnostic model is trained by the vibration signal of the first bearing in a healthy state. Therefore, the vibration signal in an abnormal state will produce a large reconstruction error in the diagnostic model, so that the normal state and the fault state can be distinguished according to the reconstruction error; in addition, the error threshold is determined according to the reconstruction error corresponding to the vibration signal in a healthy state and the vibration signal in a fault state. The error threshold can be dynamically adjusted according to the characteristics of the vibration signal at different positions, so that the error threshold is more accurate, ensuring the sensitivity and accuracy of bearing fault detection. The embodiment of the present invention effectively solves the problem that the fixed threshold in the traditional method cannot adapt to the distribution of diversified data, and improves the detection performance in different scenarios.

[0066] As an implementation manner, the error threshold is obtained by the following steps:

[0067] Obtaining a reconstruction error corresponding to each vibration signal in the second sample set according to the splicing feature corresponding to each vibration signal in the second sample set and the diagnostic model;

[0068] classifying the reconstruction error corresponding to each vibration signal in the second sample set to obtain a normal category;

[0069] The reconstruction error corresponding to the upper quartile in the normal category is used as the error threshold.

[0070] In an embodiment of the present invention, the splicing features corresponding to each vibration signal in the second sample set are input into the diagnostic model, and the reconstruction error corresponding to each vibration signal is determined according to the same method as in the above embodiment; then, the reconstruction error corresponding to each vibration signal is classified, and the reconstruction error is divided into a normal category and a fault category. For example, a K-means clustering method, a hierarchical clustering algorithm, a density clustering algorithm, a mean shift clustering algorithm, etc. may be used, and the specific method may be determined according to actual conditions, and the embodiment of the present invention does not specifically limit this. For example, in an embodiment of the present invention, a K-means clustering method is used to cluster the reconstruction error corresponding to each vibration signal, and the reconstruction error of the vibration signal is divided into a normal category and a fault category.

[0071] For the reconstruction error in the normal category, the reconstruction error corresponding to the upper quartile is taken as the error threshold. The upper quartile is also called the third quartile (Q3), which is the value at the 75% position of the data after sorting a set of data from small to large. In simple terms, the data is divided into four equal parts, and the upper quartile is the value at the starting position of the top part.

[0072] In some embodiments, the diagnostic model is an autoencoder, and the training process of the diagnostic model is as follows:

[0073] Using the vibration signal of the first bearing within a preset historical time period as the training sample;

[0074] Inputting the concatenated features corresponding to the training samples into the autoencoder to obtain the reconstructed features corresponding to the training samples;

[0075] By using a stochastic gradient descent algorithm, with the goal of minimizing the reconstruction error between the splicing features corresponding to the training samples and the reconstructed features corresponding to the training samples, the training parameters of the autoencoder are updated until the training end condition is reached, thereby obtaining the diagnostic model.

[0076] In the embodiment of the present invention, the diagnostic model can be specifically an autoencoder (AE), wherein the autoencoder is a neural network composed of an encoder and a decoder, and has a strong feature extraction capability. The AE uses its encoder to map the input sample to a low-dimensional space to obtain potential features, and then obtains the reconstructed sample through the decoder, and reconstructs the input sample with the smallest possible error during the training process.

[0077] When training the autoencoder, the vibration signal of the first bearing within a preset historical time period is first collected, and the collected vibration signal is used as a training sample. The collection steps are the same as the collection steps of the above-mentioned bearing to be tested. For details, please refer to the above embodiment, and this embodiment will not be repeated here.

[0078] The preset historical time period may be the past year, the past six months, etc., and may be determined based on actual conditions, and the embodiment of the present invention does not specifically limit this.

[0079] In the traditional method using a fixed threshold, it is necessary to deliberately collect vibration signals at different positions of the bearing in a healthy state and a faulty state and to mark them. However, in actual application, it is difficult to collect a sufficient number of vibration signals in a faulty state. In the embodiment of the present invention, it is only necessary to collect vibration signals within a preset historical time period during training, and there is no need to deliberately collect vibration signals in different states, thereby reducing the difficulty of operation.

[0080] In the embodiment of the present invention, [(X i ,y i )] represents the training sample, where i = 1, 2, ..., N, i represents the i-th detection point, N is a positive integer, N represents the total amount of vibration data contained in the vibration signal of the i-th detection point, y i is the label corresponding to the vibration signal of the i-th detection point, which can be a healthy state or a fault state.

[0081] The training sample x is then input into the autoencoder, whose encoder outputs the latent feature z, which can be expressed by the following formula:

[0082]

[0083] Among them, the network parameters of the encoder are θ e = {W e ,b e},W e and b e are the weight and bias terms of the encoder, respectively, is the activation function of the encoder.

[0084] The reconstructed features of the decoder output of the autoencoder It can be expressed as:

[0085]

[0086] Among them, the network parameters of the decoder are θ d = {W d ,b d},W d and b d are the weight and bias terms of the decoder, respectively. is the activation function of the decoder.

[0087] The ability of the autoencoder to reconstruct samples mainly depends on the learned latent features. The better the learned latent features, the stronger the ability to reconstruct samples. Then the reconstruction error is calculated based on the reconstructed features. During the training iteration, the stochastic gradient descent algorithm is used to minimize the reconstruction error during the training process, that is:

[0088]

[0089] Thereby, the training parameters of the autoencoder are updated until the training end condition is reached, and a trained autoencoder is obtained, and the trained autoencoder is used as a diagnostic model.

[0090] As an implementation manner, the vibration signal includes a plurality of vibration data, and the time domain characteristics of the vibration signal are obtained by the following steps:

[0091] According to the number of rotations of the bearing to be tested, the plurality of vibration data are divided into a plurality of vibration subsequences;

[0092] According to the time domain statistical characteristics of each vibration subsequence, the time domain characteristics corresponding to the vibration signal are obtained, wherein the time domain statistical characteristics include at least one of a mean, a variance, a skewness, a kurtosis, a coefficient of variation, a maximum amplitude, a minimum amplitude, an upper quartile, and a lower quartile.

[0093] In the embodiment of the present invention, for the i-th vibration signal X i Assuming that the vibration signal includes m×n vibration data, the vibration data in the vibration signal is divided according to the number of rotations n of the bearing to be tested to obtain n vibration data groups, each of which includes m vibration data.

[0094] The vibration data set is also called a vibration subsequence. The kth vibration subsequence corresponding to the i-th vibration signal is represented by X ik Indicates that k=1, 2, ..., n. The embodiment of the present invention calculates the mean, variance, skewness, kurtosis, coefficient of variation, maximum amplitude, minimum amplitude, upper quartile, lower quartile, etc. of each vibration subsequence corresponding to the i-th vibration signal. t time domain statistical features, and finally the time domain feature F of the i-th vibration signal is obtained s :

[0095]

[0096] in, are the time domain statistical characteristics of each vibration subsequence respectively.

[0097] As an implementation manner, the vibration signal includes a plurality of vibration data, and the frequency domain characteristics of the vibration signal are obtained by the following steps:

[0098] According to the number of rotations of the bearing to be tested, the plurality of vibration data are divided into a plurality of vibration subsequences;

[0099] After converting each vibration subsequence into the frequency domain, the power spectrum density corresponding to each vibration subsequence is calculated;

[0100] According to the frequency domain statistical characteristics of each power spectrum density, the frequency domain characteristics corresponding to the vibration signal are obtained, wherein the frequency domain statistical characteristics include at least one of average frequency, frequency variance, root mean square frequency and center frequency.

[0101] The process of obtaining the vibration subsequence in the embodiment of the present invention is the same as that in the above embodiment. For details, please refer to the above embodiment, and this embodiment will not be repeated here. Taking one of the vibration signals as an example, after obtaining the vibration subsequence corresponding to the vibration signal, each vibration subsequence is converted to frequency for spectral analysis.

[0102] In order to further analyze the characteristics of the vibration subsequence in the frequency domain, its power spectral density (PSD) is calculated to obtain the power distribution of the vibration subsequence at different frequencies. The calculation formula is as follows:

[0103]

[0104] Among them, X ik (f) represents the Fourier transform of the vibration subsequence, and m represents the length of the vibration subsequence.

[0105] Then, a series of features are extracted based on the power spectrum, namely, the average frequency Frequency variance Frequency RMS and center frequency etc. f frequency statistics characteristics.

[0106] Among them, the average frequency reflects the central trend of the frequency distribution of the vibration subsequence, the frequency variance indicates the discrete degree of the frequency distribution, the frequency root mean square reflects the energy of the frequency distribution, and the center frequency is the center point of the spectrum distribution. The calculation formula is as follows:

[0107]

[0108] Among them, f 1 is the lower limit of frequency, usually 0Hz; f 2 The upper limit of the frequency is usually half of the sampling frequency. 1 and f 2 It can be adjusted according to actual conditions.

[0109] Finally, the frequency domain feature F corresponding to the vibration signal f:

[0110]

[0111] in, are the frequency domain features corresponding to each vibration subsequence.

[0112] As an implementation manner, the step of splicing the time domain feature and the frequency domain feature to obtain the splicing feature of the vibration signal includes:

[0113] Compressing the time domain features and the frequency domain features into one-dimensional vectors respectively;

[0114] The compressed time domain features and the compressed frequency domain features are spliced ​​to obtain the spliced ​​features.

[0115] In the embodiment of the present invention, the two-dimensional time domain feature F s and frequency domain feature F f Flattening it into a one-dimensional vector gives:

[0116]

[0117] At this time, F′ s is of length n×c t The time domain eigenvector of f ′ is of length n×c f The time domain feature vector of .

[0118] Final splicing F′ s and F′ f , get the i-th vibration signal X i The splicing features Its length is l = n × c t +n×c f .

[0119] The time domain features and frequency domain features of all vibration signals are extracted and merged to obtain the feature data set TF:

[0120] TF=[F 1 F 2 … F N ] T .

[0121] Among them, the size of TF is n×l, and the i-th row of data represents the vibration signal X i splicing features.

[0122] In some embodiments, before the step of inputting the splicing features into the diagnostic model to obtain the reconstruction features, the method further comprises:

[0123] The splicing features are normalized.

[0124] In the embodiment of the present invention, before the splicing features are input into the diagnostic model, the splicing features need to be normalized. Since different units of characteristic indicators have different degrees of influence on the bearing operation status detection results, the splicing features with larger scales will play a decisive role, while the splicing features with smaller scales may be ignored. In order to eliminate the influence of the unit and scale differences between these features, they need to be normalized, that is, the feature data is mapped to the [0,1] interval. The calculation formula is as follows:

[0125]

[0126] This method can achieve proportional scaling of the original feature index. norm is the normalized concatenated feature, TF is the original concatenated feature, and TF max TF min are the maximum and minimum values ​​of all splicing features respectively.

[0127] Figure 2 A structural diagram of a bearing fault diagnosis device provided by an embodiment of the present invention, such as Figure 2 As shown, the device comprises:

[0128] The acquisition module 210 is used to acquire the vibration signal of the bearing to be tested, the time domain characteristics of the vibration signal, and the frequency domain characteristics of the vibration signal;

[0129] A splicing module 220, configured to splice the time domain feature and the frequency domain feature to obtain a splicing feature of the vibration signal;

[0130] A reconstruction module 230, configured to input the spliced ​​features into a diagnostic model to obtain a reconstructed feature, wherein the diagnostic model is obtained by training a first sample set, the first sample set includes training samples, and the training samples are vibration signals of the first bearing in a healthy state;

[0131] The diagnostic module 240 is used to determine whether the bearing to be tested has a fault based on a reconstruction error between the reconstructed feature and the spliced ​​feature and an error threshold, wherein the error threshold is determined based on a reconstruction error generated after a second sample set is input into the diagnostic model, and the second sample set includes a vibration signal of the second bearing in a healthy state and a vibration signal of the third bearing in a faulty state.

[0132] This embodiment is a device embodiment corresponding to the above method, and the specific implementation process is the same as that of the above method embodiment. For details, please refer to the above method embodiment, and this device embodiment will not be repeated here.

[0133] Based on the above bearing fault diagnosis method, an embodiment of the present invention further provides a computer device, Figure 3 A schematic diagram of the structure of a computer device provided by an embodiment of the present invention is shown in FIG. Figure 3 As shown, the computer device includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0134] Memory, used to store computer programs;

[0135] The processor is used to implement the steps of the above-mentioned task scheduling processing method when executing the program stored in the memory.

[0136] For other details about the processor in the above-mentioned computer device implementing the above-mentioned technical solution, please refer to the description of the task scheduling processing method provided in the above-mentioned invention embodiment, which will not be repeated here.

[0137] Among them, the processor can also be called CPU (Central Processing Unit), the processor may be an integrated circuit chip with signal processing capabilities; the processor can also be a general-purpose processor, DSP (Digital Signal Process), ASIC (Application Specific Integrated Circuit), FPGA (Field Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, among which the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0138] The embodiment of the present invention also provides a computer-readable storage medium, on which a readable computer program is stored; wherein the computer program can be stored in the above storage medium in the form of a software product, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, magnetic disk or optical disk, ROM (Read-Only Memory), RAM (Random Access Memory) and other media that can store program codes, or terminal devices such as computers, servers, mobile phones, and tablets.

[0139] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules 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 an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0140] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0141] In addition, each functional module in each embodiment of the present application can be integrated into a processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above-mentioned integrated module can be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0142] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0143] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, a computer, a server, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server, or data center. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or a data center that includes one or more available media integrations. The available medium may be a magnetic medium, (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0144] The technical solution provided by the present application is introduced in detail above. The principles and implementation methods of the present application are explained by using specific examples in the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

[0145] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0146] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0147] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0148] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0149] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A bearing fault diagnosis method, characterized in that: The method comprises: Acquire a vibration signal of the bearing to be tested, a time domain feature of the vibration signal, and a frequency domain feature of the vibration signal; Splicing the time domain features and the frequency domain features to obtain a splicing feature of the vibration signal; Inputting the spliced ​​features into a diagnostic model to obtain a reconstructed feature, wherein the diagnostic model is obtained by training a first sample set, the first sample set includes training samples, and the training samples are vibration signals of the first bearing in a healthy state; Based on the reconstruction error between the reconstructed feature and the splicing feature and the error threshold, it is determined whether the bearing to be tested is faulty, wherein the error threshold is determined based on the reconstruction error generated after the second sample set is input into the diagnostic model, and the second sample set includes the vibration signal of the second bearing in a healthy state and the vibration signal of the third bearing in a faulty state.

2. The bearing fault diagnosis method according to claim 1, characterized in that: The error threshold is obtained by the following steps: Obtaining a reconstruction error corresponding to each vibration signal in the second sample set according to the splicing feature corresponding to each vibration signal in the second sample set and the diagnostic model; classifying the reconstruction error corresponding to each vibration signal in the second sample set to obtain a normal category; The reconstruction error corresponding to the upper quartile in the normal category is used as the error threshold.

3. The bearing fault diagnosis method according to claim 1, characterized in that: The diagnostic model is an autoencoder, and the training process of the diagnostic model is as follows: Using the vibration signal of the first bearing within a preset historical time period as the training sample; Inputting the concatenated features corresponding to the training samples into the autoencoder to obtain the reconstructed features corresponding to the training samples; By using a stochastic gradient descent algorithm, with the goal of minimizing the reconstruction error between the splicing features corresponding to the training samples and the reconstructed features corresponding to the training samples, the training parameters of the autoencoder are updated until the training end condition is reached, thereby obtaining the diagnostic model.

4. The bearing fault diagnosis method according to claim 1, characterized in that: The vibration signal includes a plurality of vibration data, and the time domain characteristics of the vibration signal are obtained by the following steps: According to the number of rotations of the bearing to be tested, the plurality of vibration data are divided into a plurality of vibration subsequences; According to the time domain statistical characteristics of each vibration subsequence, the time domain characteristics corresponding to the vibration signal are obtained, wherein the time domain statistical characteristics include at least one of a mean, a variance, a skewness, a kurtosis, a coefficient of variation, a maximum amplitude, a minimum amplitude, an upper quartile, and a lower quartile.

5. The bearing fault diagnosis method according to claim 1, characterized in that: The vibration signal includes a plurality of vibration data, and the frequency domain characteristics of the vibration signal are obtained by the following steps: According to the number of rotations of the bearing to be tested, the plurality of vibration data are divided into a plurality of vibration subsequences; After converting each vibration subsequence into the frequency domain, the power spectrum density corresponding to each vibration subsequence is calculated; According to the frequency domain statistical characteristics of each power spectrum density, the frequency domain characteristics corresponding to the vibration signal are obtained, wherein the frequency domain statistical characteristics include at least one of average frequency, frequency variance, root mean square frequency and center frequency.

6. The bearing fault diagnosis method according to any one of claims 1 to 5, characterized in that: The step of splicing the time domain feature and the frequency domain feature to obtain the splicing feature of the vibration signal comprises: Compressing the time domain features and the frequency domain features into one-dimensional vectors respectively; The compressed time domain features and the compressed frequency domain features are spliced ​​to obtain the spliced ​​features.

7. The bearing fault diagnosis method according to any one of claims 1 to 5, characterized in that: Before the step of inputting the splicing features into the diagnosis model to obtain the reconstruction features, the method further comprises: The splicing features are normalized.

8. A bearing fault diagnosis device, characterized in that: The device comprises: An acquisition module, used to acquire the vibration signal of the bearing to be tested, the time domain characteristics of the vibration signal, and the frequency domain characteristics of the vibration signal; A splicing module, used for splicing the time domain features and the frequency domain features to obtain a splicing feature of the vibration signal; A reconstruction module, used for inputting the splicing features into a diagnostic model to obtain a reconstructed feature, wherein the diagnostic model is obtained by training a first sample set, the first sample set includes training samples, and the training samples are vibration signals of the first bearing in a healthy state; A diagnostic module is used to determine whether the bearing to be tested has a fault based on a reconstruction error between the reconstructed feature and the spliced ​​feature and an error threshold, wherein the error threshold is determined based on a reconstruction error generated after a second sample set is input into the diagnostic model, and the second sample set includes a vibration signal of the second bearing in a healthy state and a vibration signal of the third bearing in a faulty state.

9. A computer device, characterized in that: The device includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; Memory, used to store computer programs; The processor is used to implement the steps of the bearing fault diagnosis method described in any one of claims 1 to 7 when executing the program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the bearing fault diagnosis method according to any one of claims 1 to 7 are implemented.

Images