Axle box bearing fault diagnosis method and system

Abstract

The invention provides an axle box bearing fault diagnosis method and system. The method comprises the steps of performing normalization processing on a bearing fault signal data set; gaussian noise is introduced, and forward diffusion processing is carried out on the normalized data set through the Gaussian noise; inputting the forward diffusion data set into a training backward diffusion model, and performing backward diffusion processing on the forward diffusion data set according to predicted noise output by the training backward diffusion model; and inputting the feature data into a preset diagnosis model for training, obtaining a to-be-diagnosed bearing data set, inputting the to-be-diagnosed bearing data set into the trained preset diagnosis model for fault diagnosis, and outputting a diagnosis result. And moreover, various different working conditions and fault types can be simulated, the data diversity is improved, and meanwhile, the fault diagnosis precision can also be improved.

Description

Technical Field

[0001] The invention belongs to the technical field of fault diagnosis, and in particular relates to an axle box bearing fault diagnosis method and system. Background Art

[0002] Axlebox bearings are core components of high-speed trains, and their operating status directly affects the performance of the entire train. Common failure modes of axlebox bearings include inner ring, outer ring and rolling element failures, and each failure will produce characteristic changes in its corresponding vibration signal.

[0003] In recent years, in order to improve the accuracy of fault diagnosis, machine learning and deep learning technologies have gradually become research hotspots in the field of bearing fault diagnosis. In particular, methods such as convolutional neural networks (CNN), recurrent neural networks (RNN), and deep autoencoders (AE) have achieved remarkable results in bearing fault diagnosis. However, these methods usually rely on a large amount of labeled data for training, and in practical applications, it is often difficult to obtain sufficient fault sample data. Especially in the case of special fault modes or rare fault types, the existing fault sample data is seriously insufficient, which limits the training effect of deep learning models and the accuracy of fault diagnosis.

[0004] In order to solve this problem, data enhancement technology has received widespread attention as an effective supplementary means. Traditional data enhancement methods, such as those based on image flipping, scaling, and cropping, are mainly used in the field of image recognition. However, the application of existing data enhancement methods in bearing fault diagnosis still faces challenges. Especially in the case of high noise and non-stationary signals, how to effectively generate high-quality and representative fault data is still a difficult problem that needs to be solved urgently. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides an axle box bearing fault diagnosis method and system for solving the technical problems in the prior art.

[0006] On the one hand, the present invention provides the following technical solution, a method for diagnosing an axle box bearing fault, comprising: Acquire a time-continuous bearing fault signal data set, and perform normalization processing on the bearing fault signal data set to obtain a normalized data set; Introducing Gaussian noise and performing forward diffusion processing on the normalized data set by the Gaussian noise to obtain a forward diffusion data set; The reverse diffusion model is trained by the bearing fault signal data set to obtain a trained reverse diffusion model, the forward diffusion data set is input into the trained reverse diffusion model, and the forward diffusion data set is subjected to reverse diffusion processing according to the predicted noise output by the trained reverse diffusion model to obtain a reverse diffusion data set; Extract characteristic data of the back diffusion data set and the bearing fault signal data set, input the characteristic data into a preset diagnostic model for training, obtain the bearing data set to be diagnosed, input the bearing data set to be diagnosed into the trained preset diagnostic model for fault diagnosis, and output a diagnostic result.

[0007] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention first obtains a time-continuous bearing fault signal data set, normalizes the bearing fault signal data set to obtain a normalized data set; then introduces Gaussian noise and performs forward diffusion processing on the normalized data set through the Gaussian noise to obtain a forward diffusion data set; then trains a reverse diffusion model through the bearing fault signal data set to obtain a trained reverse diffusion model, inputs the forward diffusion data set into the trained reverse diffusion model, and performs reverse diffusion processing on the forward diffusion data set according to the predicted noise output by the trained reverse diffusion model to obtain a reverse diffusion data set; then extracts feature data of the reverse diffusion data set and the bearing fault signal data set, inputs the feature data into a preset diagnostic model for training, obtains a bearing data set to be diagnosed, inputs the bearing data set to be diagnosed into the trained preset diagnostic model for fault diagnosis, and outputs a diagnostic result. The present invention generates data through a reverse diffusion model, and the generated data is not only close to the original data in statistical characteristics, but also can simulate a variety of different working conditions and fault types, improves the diversity of data, and also can improve the diagnostic accuracy of fault diagnosis.

[0008] Preferably, the step of normalizing the bearing fault signal data set to obtain a normalized data set includes: The bearing fault signal data set is normalized using a normalization formula to obtain a normalized data set, wherein the normalization formula is: ; In the formula, , They represent the bearing fault signal dataset and the normalized dataset. Signal data, They respectively represent the maximum and minimum signal data in the bearing fault signal data set.

[0009] Preferably, the step of introducing Gaussian noise and performing forward diffusion processing on the normalized data set by the Gaussian noise to obtain the forward diffusion data set includes: Determine the strength factor : ; In the formula, represents the hyperparameter, Indicates time steps; Based on the strength coefficient With Gaussian noise Determine the forward diffusion data set : ; ; In the formula, , Indicates the forward diffusion data set , signal data set, Indicates that the mean is 0 and the covariance matrix is ​​the unit matrix No. Gaussian noise corresponding to the time step.

[0010] Preferably, the step of training the reverse diffusion model by using the bearing fault signal data set to obtain the trained reverse diffusion model includes: The bearing fault signal data set is input into the U-Net network for noise prediction to obtain the predicted noise ; Based on the predicted noise Determine the loss function : ; In the formula, represents the mathematical expectation, represents the actual noise; The loss function is minimized to train the reverse diffusion model to obtain a trained reverse diffusion model.

[0011] Preferably, the step of performing reverse diffusion processing on the forward diffusion data set according to the predicted noise output by the trained reverse diffusion model to obtain the reverse diffusion data set comprises: The signal data set of the last time step in the forward diffusion data set Diffusion processing is performed to obtain the reverse diffusion data set :

[0012] ; ; In the formula, , Represents the diffusion data group Middle , signal data set, Respectively represent , , The intensity coefficient corresponding to the time step is, , They represent the output of the training back-diffusion model. , The noise component corresponding to the time step is , They represent the output of the training back-diffusion model. , The prediction noise corresponding to the time step; The diffusion data set The first signal data set as a back-diffusion dataset.

[0013] Preferably, the characteristic data includes signal energy, main frequency, and envelope analysis of the back diffusion data set and the bearing fault signal data set.

[0014] Preferably, the preset diagnostic model is specifically a CNN model.

[0015] In a second aspect, the present invention provides the following technical solution, a system for diagnosing axle box bearing fault, the system comprising: A normalization module, used to obtain a time-continuous bearing fault signal data set, and perform normalization processing on the bearing fault signal data set to obtain a normalized data set; A forward module, used for introducing Gaussian noise and performing forward diffusion processing on the normalized data set by the Gaussian noise to obtain a forward diffusion data set; A reverse module, used for training a reverse diffusion model with the bearing fault signal data set to obtain a trained reverse diffusion model, inputting the forward diffusion data set into the trained reverse diffusion model, and performing reverse diffusion processing on the forward diffusion data set according to the predicted noise output by the trained reverse diffusion model to obtain a reverse diffusion data set; A diagnostic module is used to extract feature data of the back diffusion data set and the bearing fault signal data set, input the feature data into a preset diagnostic model for training, obtain a bearing data set to be diagnosed, input the bearing data set to be diagnosed into the trained preset diagnostic model for fault diagnosis, and output a diagnostic result.

[0016] In a third aspect, the present invention provides the following technical solution: a computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the axle box bearing fault diagnosis method as described above when executing the computer program.

[0017] In a fourth aspect, the present invention provides the following technical solution: a storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the axle box bearing fault diagnosis method as described above. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, 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 invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flow chart of the axle box bearing fault diagnosis method provided in the first embodiment of the present invention; Figure 2 A model structure diagram of the reverse diffusion model provided in the first embodiment of the present invention; Figure 3 This is a frequency domain comparison diagram of the back diffusion data set and the bearing fault signal data set provided in the first embodiment of the present invention.

[0020] Figure 4 A structural block diagram of an axle box bearing fault diagnosis system provided in Embodiment 2 of the present invention; Figure 5 A schematic diagram of the hardware structure of a computer provided in another embodiment of the present invention.

[0021] The embodiments of the present invention will be further described below with reference to the accompanying drawings. DETAILED DESCRIPTION

[0022] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the embodiments of the present invention, and should not be construed as limiting the present invention.

[0023] In the description of the embodiments of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside" and "outside" etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the embodiments of the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present invention.

[0024] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0025] In the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the embodiments of the present invention can be understood according to specific circumstances.

[0026] Embodiment 1 In the first embodiment of the present invention, Figure 1 As shown, a method for diagnosing axle box bearing fault includes: S1. Acquire a time-continuous bearing fault signal data set, and perform normalization processing on the bearing fault signal data set to obtain a normalized data set; Wherein, step S1 comprises: The bearing fault signal data set is normalized using a normalization formula to obtain a normalized data set, wherein the normalization formula is: ; In the formula, , They represent the bearing fault signal dataset and the normalized dataset. Signal data, Respectively represent the maximum and minimum signal data in the bearing fault signal data set; Among them, the data in the bearing fault signal data set can be obtained through sensors, such as vibration sensors, temperature sensors, etc., which record the operating data of the equipment under normal and fault conditions. The collected data includes different types of bearing fault samples, such as inner ring fault, outer ring fault, rolling element fault, etc.

[0027] S2, introducing Gaussian noise and performing forward diffusion processing on the normalized data set by the Gaussian noise to obtain a forward diffusion data set; Wherein, the step S2 comprises: S21. Determine the strength coefficient : ; In the formula, represents the hyperparameter, Indicates time steps.

[0028] S22, based on the strength coefficient With Gaussian noise Determine the forward diffusion data set : ; ; In the formula, , Indicates the forward diffusion data set , signal data set, Indicates that the mean is 0 and the covariance matrix is ​​the unit matrix No. Gaussian noise corresponding to time steps; Specifically, in the process of forward diffusion, the core is to gradually add noise to the signal to simulate the actual noise interference. The diffusion model gradually increases the noise through multiple time steps, but the increase of noise is not linear, but follows a certain probability distribution, which can better simulate the noise changes under complex working conditions. The forward diffusion process is usually achieved by introducing Gaussian noise, and for the introduced Gaussian noise, its intensity is controlled by the intensity coefficient. Therefore, in actual situations, it is hoped that in the process of forward diffusion, the signal will be gradually covered by noise. Therefore, in the process of forward diffusion, as the diffusion steps proceed, When increasing, Gradually decrease, the noise component It will gradually increase, and the noise will gradually dominate the signal; In the diffusion process, the diffusion process is divided into multiple time steps in a multi-level diffusion manner. Each time step corresponds to a different noise coefficient. By gradually increasing the noise, samples close to Gaussian noise can eventually be obtained. These samples will be used for denoising in the subsequent reverse diffusion process.

[0029] S3, training a reverse diffusion model using the bearing fault signal data set to obtain a trained reverse diffusion model, inputting the forward diffusion data set into the trained reverse diffusion model, and performing reverse diffusion processing on the forward diffusion data set according to the predicted noise output by the trained reverse diffusion model to obtain a reverse diffusion data set; The step of training the reverse diffusion model by using the bearing fault signal data set to obtain the trained reverse diffusion model includes: S311, inputting the bearing fault signal data set into the U-Net network for noise prediction to obtain predicted noise .

[0030] S312, based on the predicted noise Determine the loss function : ; In the formula, represents the mathematical expectation, Represents the actual noise.

[0031] S313, performing minimization processing on the loss function to train the reverse diffusion model to obtain a trained reverse diffusion model; Specifically, the U-Net network can predict the noise at each step by learning the diffusion process in the bearing fault signal data set. The reverse diffusion model can be trained to learn how to accurately estimate the noise component from the noisy signal, thereby gradually removing the noise in the reverse diffusion process. After training, data with bearing fault characteristics can be generated. The structure diagram of the U-Net network is shown in the figure. Figure 2 shown.

[0032] The step of performing reverse diffusion processing on the forward diffusion data set according to the predicted noise output by the trained reverse diffusion model to obtain the reverse diffusion data set includes: S321, the signal data set of the last time step in the forward diffusion data set Diffusion processing is performed to obtain the reverse diffusion data set :

[0033] ; ; In the formula, , Represents the diffusion data group Middle , signal data set, Respectively represent , , The intensity coefficient corresponding to the time step is, , They represent the output of the training back-diffusion model. , The noise component corresponding to the time step is , They represent the output of the training back-diffusion model. , The prediction noise corresponding to the time step; S322, the reverse diffusion data group The first signal data set As a back-diffusion dataset; Specifically, the signal data set The sample data is closest to Gaussian noise, so it is used as the input of the model from a signal data set with a large amount of noise signal. , gradually recovering through denoising until it is close to the original noise-free signal data set .

[0034] It should be noted that if Figure 3 As shown, the original signal is the data signal in the bearing fault signal dataset, and the generated signal is the data signal in the reverse diffusion dataset. The collected fault data is used to train the reverse diffusion model. The reverse diffusion model is a generative model that can generate samples similar to real fault data by learning the distribution and denoising process of noise data. During the training process, the model will learn how to recover the potential structure and characteristics of the fault data from the noise data, and through the trained reverse diffusion model, new data can be generated. The data has similar characteristics to the original fault data and can effectively supplement the data set of the original samples. The generated data is not only close to the original data in statistical characteristics, but also can simulate a variety of different working conditions and fault types to improve data diversity.

[0035] S4. Extract feature data of the back diffusion data set and the bearing fault signal data set, input the feature data into a preset diagnostic model for training, obtain a bearing data set to be diagnosed, input the bearing data set to be diagnosed into the trained preset diagnostic model for fault diagnosis, and output a diagnostic result.

[0036] Among them, the characteristic data includes signal energy, main frequency, and envelope analysis of the back diffusion data set and the bearing fault signal data set, and the preset diagnostic model is specifically a CNN model.

[0037] Specifically, by extracting feature data from the back-diffusion dataset and the bearing fault signal dataset, the back-diffusion dataset can expand the bearing fault signal dataset. As the number of data samples increases, subsequent fault diagnosis models can be trained on richer samples, thereby improving the generalization ability and diagnostic accuracy of the model. At the same time, the generated samples can be combined with real data and mixed in a certain proportion. The trained CNN model can accurately diagnose bearing faults under different working conditions and improve the diagnostic accuracy under complex working conditions, ensuring the diversity and representativeness of the data. The trained CNN model can perform fault diagnosis on the bearing dataset to be diagnosed and output the corresponding diagnostic results.

[0038] The axle box bearing fault diagnosis method provided in the first embodiment of the present invention first obtains a time-continuous bearing fault signal data set, normalizes the bearing fault signal data set to obtain a normalized data set; then introduces Gaussian noise and performs forward diffusion processing on the normalized data set through the Gaussian noise to obtain a forward diffusion data set; then trains a reverse diffusion model through the bearing fault signal data set to obtain a trained reverse diffusion model, inputs the forward diffusion data set into the trained reverse diffusion model, and performs reverse diffusion processing on the forward diffusion data set according to the predicted noise output by the trained reverse diffusion model to obtain a reverse diffusion data set; then extracts feature data of the reverse diffusion data set and the bearing fault signal data set, inputs the feature data into a preset diagnosis model for training, obtains a bearing data set to be diagnosed, inputs the bearing data set to be diagnosed into the trained preset diagnosis model for fault diagnosis, and outputs a diagnosis result. The present invention generates data through a reverse diffusion model, and the generated data is not only close to the original data in statistical characteristics, but also can simulate a variety of different working conditions and fault types, improves the diversity of data, and also can improve the diagnostic accuracy of fault diagnosis.

[0039] Embodiment 2 like Figure 4 As shown, in the second embodiment of the present invention, a system for diagnosing axle box bearing fault is provided, and the system comprises: A normalization module 1 is used to obtain a time-continuous bearing fault signal data set, and perform normalization processing on the bearing fault signal data set to obtain a normalized data set; A forward module 2, used for introducing Gaussian noise and performing forward diffusion processing on the normalized data set by the Gaussian noise to obtain a forward diffusion data set; A reverse module 3 is used to train a reverse diffusion model using the bearing fault signal data set to obtain a trained reverse diffusion model, input the forward diffusion data set into the trained reverse diffusion model, and perform reverse diffusion processing on the forward diffusion data set according to the predicted noise output by the trained reverse diffusion model to obtain a reverse diffusion data set; Diagnostic module 4, used for extracting characteristic data of the back diffusion data set and the bearing fault signal data set, inputting the characteristic data into a preset diagnostic model for training, obtaining a bearing data set to be diagnosed, inputting the bearing data set to be diagnosed into the trained preset diagnostic model for fault diagnosis, and outputting a diagnostic result; The normalization module 1 is specifically used for: The bearing fault signal data set is normalized using a normalization formula to obtain a normalized data set, wherein the normalization formula is: ; In the formula, , They represent the bearing fault signal dataset and the normalized dataset. Signal data, They respectively represent the maximum and minimum signal data in the bearing fault signal data set.

[0040] The forward module 2 comprises: Coefficient submodule, used to determine the strength coefficient : ; In the formula, represents the hyperparameter, Indicates time steps; The forward diffusion submodule is used to With Gaussian noise Determine the forward diffusion data set : ; ; In the formula, , Indicates the forward diffusion data set , signal data set, Indicates that the mean is 0 and the covariance matrix is ​​the unit matrix No. Gaussian noise corresponding to the time step.

[0041] The reverse module 3 comprises: The prediction submodule is used to input the bearing fault signal data set into the U-Net network for noise prediction to obtain the predicted noise ; Function submodule for predicting noise based on the Determine the loss function : ; In the formula, represents the mathematical expectation, represents the actual noise; The training submodule is used to minimize the loss function to train the reverse diffusion model to obtain a trained reverse diffusion model.

[0042] The reverse module 3 also includes: The reverse diffusion submodule is used to convert the signal data set of the last time step in the forward diffusion data set into Diffusion processing is performed to obtain the reverse diffusion data set :

[0043] ; ; In the formula, , Represents the diffusion data group Middle , signal data set, Respectively represent , , The intensity coefficient corresponding to the time step is, , They represent the output of the training back-diffusion model. , The noise component corresponding to the time step is , They represent the output of the training back-diffusion model. , The prediction noise corresponding to the time step; An output submodule is used to output the reverse diffusion data set The first signal data set as a back-diffusion dataset.

[0044] In some other embodiments of the present invention, the embodiments of the present invention provide the following technical solution: a computer, comprising a memory 102, a processor 101, and a computer program stored in the memory 102 and executable on the processor 101, wherein the processor 101 implements the axle box bearing fault diagnosis method as described above when executing the computer program.

[0045] Specifically, the processor 101 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiment of the present invention.

[0046] Among them, the memory 102 may include a large-capacity memory for data or instructions. By way of example and not limitation, the memory 102 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 102 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 102 may be inside or outside the data processing device. In a specific embodiment, the memory 102 is a non-volatile memory. In a specific embodiment, the memory 102 includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM) or a flash memory (FLASH), or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), wherein the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0047] The memory 102 may be used to store or cache various data files that need to be processed and / or used for communication, as well as possible computer program instructions executed by the processor 101 .

[0048] The processor 101 implements the above-mentioned axle box bearing fault diagnosis method by reading and executing the computer program instructions stored in the memory 102 .

[0049] In some embodiments, the computer may further include a communication interface 103 and a bus 100. Figure 5 As shown, the processor 101, the memory 102, and the communication interface 103 are connected via a bus 100 and communicate with each other.

[0050] The communication interface 103 is used to implement communication between the modules, devices, units and / or equipment in the embodiment of the present invention. The communication interface 103 can also implement data communication with other components such as: external devices, image / data acquisition equipment, databases, external storage, and image / data processing workstations.

[0051] The bus 100 includes hardware, software or both, and couples the components of the computer device to each other. The bus 100 includes but is not limited to at least one of the following: a data bus, an address bus, a control bus, an expansion bus, and a local bus. By way of example and not limitation, bus 100 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, bus 100 may include one or more buses. Although embodiments of the present invention describe and illustrate a particular bus, the present invention contemplates any suitable bus or interconnect.

[0052] The computer can execute the axle box bearing fault diagnosis method of the present invention based on the acquired axle box bearing fault diagnosis system, thereby realizing axle box bearing fault diagnosis.

[0053] In some further embodiments of the present invention, in combination with the above-mentioned axle box bearing fault diagnosis method, the embodiments of the present invention provide the following technical solutions: a storage medium having a computer program stored thereon, and the computer program implements the above-mentioned axle box bearing fault diagnosis method when executed by a processor.

[0054] Those skilled in the art will appreciate that the logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For purposes of this specification, "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0055] More specific examples of readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0056] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0057] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0058] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A method for diagnosing axle box bearing fault, characterized in that: include: Acquire a time-continuous bearing fault signal data set, and perform normalization processing on the bearing fault signal data set to obtain a normalized data set; Introducing Gaussian noise and performing forward diffusion processing on the normalized data set by the Gaussian noise to obtain a forward diffusion data set; The reverse diffusion model is trained by the bearing fault signal data set to obtain a trained reverse diffusion model, the forward diffusion data set is input into the trained reverse diffusion model, and the forward diffusion data set is subjected to reverse diffusion processing according to the predicted noise output by the trained reverse diffusion model to obtain a reverse diffusion data set; Extract characteristic data of the back diffusion data set and the bearing fault signal data set, input the characteristic data into a preset diagnostic model for training, obtain the bearing data set to be diagnosed, input the bearing data set to be diagnosed into the trained preset diagnostic model for fault diagnosis, and output a diagnostic result.

2. The axle box bearing fault diagnosis method according to claim 1, characterized in that: The step of normalizing the bearing fault signal data set to obtain a normalized data set includes: The bearing fault signal data set is normalized using a normalization formula to obtain a normalized data set, wherein the normalization formula is: ； In the formula, , They represent the bearing fault signal dataset and the normalized dataset. Signal data, They respectively represent the maximum and minimum signal data in the bearing fault signal data set.

3. The axle box bearing fault diagnosis method according to claim 1, characterized in that: The step of introducing Gaussian noise and performing forward diffusion processing on the normalized data set by the Gaussian noise to obtain the forward diffusion data set includes: Determine the strength factor : ； In the formula, represents the hyperparameter, Indicates time steps; Based on the strength coefficient With Gaussian noise Determine the forward diffusion data set : ； ； In the formula, , Indicates the forward diffusion data set , signal data set, Indicates that the mean is 0 and the covariance matrix is ​​the unit matrix No. Gaussian noise corresponding to the time step.

4. The axle box bearing fault diagnosis method according to claim 1, characterized in that: The step of training the reverse diffusion model by using the bearing fault signal data set to obtain the trained reverse diffusion model comprises: The bearing fault signal data set is input into the U-Net network for noise prediction to obtain the predicted noise ; Based on the predicted noise Determine the loss function : ； In the formula, represents the mathematical expectation, represents the actual noise; The loss function is minimized to train the reverse diffusion model to obtain a trained reverse diffusion model.

5. The axle box bearing fault diagnosis method according to claim 1, characterized in that: The step of performing reverse diffusion processing on the forward diffusion data set according to the predicted noise output by the trained reverse diffusion model to obtain the reverse diffusion data set comprises: The signal data set of the last time step in the forward diffusion data set Diffusion processing is performed to obtain the reverse diffusion data set : ； ； In the formula, , Represents the diffusion data group Middle , signal data set, Respectively represent , , The intensity coefficient corresponding to the time step is , They represent the output of the training back-diffusion model. , The noise component corresponding to the time step is , They represent the output of the training back-diffusion model. , The prediction noise corresponding to the time step; The back diffusion data set The first signal data set as a back-diffusion dataset.

6. The axle box bearing fault diagnosis method according to claim 1, characterized in that: The characteristic data includes signal energy, main frequency, and envelope analysis of the back diffusion data set and the bearing fault signal data set.

7. The axle box bearing fault diagnosis method according to claim 1, characterized in that: The preset diagnostic model is specifically a CNN model.

8. An axle box bearing fault diagnosis system, characterized in that: The system comprises: A normalization module, used to obtain a time-continuous bearing fault signal data set, and perform normalization processing on the bearing fault signal data set to obtain a normalized data set; A forward module, used for introducing Gaussian noise and performing forward diffusion processing on the normalized data set by the Gaussian noise to obtain a forward diffusion data set; A reverse module, used for training a reverse diffusion model with the bearing fault signal data set to obtain a trained reverse diffusion model, inputting the forward diffusion data set into the trained reverse diffusion model, and performing reverse diffusion processing on the forward diffusion data set according to the predicted noise output by the trained reverse diffusion model to obtain a reverse diffusion data set; A diagnostic module is used to extract feature data of the back diffusion data set and the bearing fault signal data set, input the feature data into a preset diagnostic model for training, obtain a bearing data set to be diagnosed, input the bearing data set to be diagnosed into the trained preset diagnostic model for fault diagnosis, and output a diagnostic result.

9. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the axle box bearing fault diagnosis method according to any one of claims 1 to 7 is implemented.

10. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by the processor, the axle box bearing fault diagnosis method according to any one of claims 1 to 7 is implemented.

Images