Bearing fault diagnosis method, equipment, cluster, medium, program product and device

Abstract

The invention discloses a bearing fault diagnosis method, equipment, a cluster, a medium, a program product and a device, relates to the technical field of high-speed train bogie bearing fault diagnosis, provides a bearing fault diagnosis method based on minimization of Shannon entropy and information enhancement, and mainly solves the problems that a current bogie bearing fault model of a high-speed train is low in bogie crossing accuracy rate and high in reliability. Generalizing the problem; the method comprises the following steps: acquiring fault information of a bearing; designing a convolutional neural network including Shannon entropy calculation and bearing signal feature enhancement; optimizing a filtering convolution layer through minimizing the Shannon entropy; carrying out feature enhancement on the bearing fault features; outputting a fault diagnosis result; and carrying out migration diagnosis in the cross-equipment field. Through the bearing fault diagnosis method and device, the problem that the field generalization accuracy of bearing fault diagnosis under the time-varying working condition is low is solved, and the accuracy and reliability of bearing fault diagnosis are further improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of bearing fault diagnosis, and more specifically, relates to a bearing fault diagnosis method, equipment, cluster, medium, program product and device. Background Art

[0002] High-speed trains are an essential component of modern transportation, and their safe and stable operation is crucial. The health of bogie bearings, critical load-bearing components, directly impacts train safety. Therefore, efficient and accurate bearing fault diagnosis technology is crucial for preventing train accidents and ensuring passenger safety. Currently, with increasing train speeds and increasingly complex operating environments, bearings may encounter a variety of operating conditions and loads during operation, presenting new challenges for bearing condition monitoring and fault diagnosis.

[0003] Existing bearing fault diagnosis technologies are mostly based on traditional signal processing and machine learning methods. While these methods can achieve high diagnostic accuracy under specific conditions, they often rely on large amounts of labeled data and stable operating conditions. However, in practical applications, labeled data is difficult to obtain, and bearing operating conditions frequently change, severely impacting the generalization and practicality of existing technologies. Furthermore, the nonlinear and non-stationary nature of bearing fault signals makes it difficult for traditional methods to accurately capture fault characteristics, thus affecting the accuracy and reliability of fault diagnosis. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to address the deficiencies in the above-mentioned prior art and provide a bearing fault diagnosis method, equipment, cluster, medium, program product and device, which improves the accuracy and generalization ability of fault diagnosis by minimizing Shannon entropy and information enhancement, and is used to solve the current technical problem that it is difficult to accurately capture fault characteristics, thereby affecting the accuracy and reliability of fault diagnosis.

[0005] The present invention adopts the following technical solutions: In a first aspect, a bearing fault diagnosis method is provided, comprising the following steps: Obtain the original fault signal of the bearing; Design a convolutional neural network that includes Shannon entropy calculation and bearing signal feature enhancement, construct a filtering convolution layer, and optimize the parameters of the convolutional neural network by minimizing the Shannon entropy of the signal after passing through the filtering layer; Calculate the bearing fault frequency and build a feature enhancement layer to enhance the bearing fault features; Construct traditional convolutional layers as classifier networks for fault diagnosis, output fault diagnosis results, and perform migration diagnosis across device domains.

[0006] Preferably, a convolutional neural network is designed that includes Shannon entropy calculation and bearing signal feature enhancement, a filter convolution layer is constructed, and the parameters of the convolutional neural network are optimized by minimizing the Shannon entropy of the signal after passing through the filter layer, specifically: Construct a filtering convolution layer, which calculates the weighted sum of the local area by sliding convolution on the input data and performs filtering on the signal by constraining the convolution kernel parameters; Input the vibration signal data into the filter convolution layer and calculate the Shannon entropy of the output signal; The signal of the filtered convolution layer optimized by minimizing Shannon entropy is input into the subsequent feature extraction and classification network. The Shannon entropy minimization optimization process is carried out separately from the subsequent model feature extraction process.

[0007] Preferably, a feature enhancement layer is constructed to enhance the bearing fault features, specifically: The characteristic frequency of the bearing is calculated based on the bearing size and shape data; a feature enhancement layer is designed to enhance the signal characteristics using the frequency characteristics of the bearing rotation frequency, inner ring fault frequency, outer ring fault frequency, and roller fault frequency.

[0008] Preferably, a traditional convolutional layer is constructed as a classifier network for fault diagnosis, the fault diagnosis results are output, and migration diagnosis is performed across device domains, specifically: Vibration data from two different bogies are obtained, with the data from one bogie as the source domain and the data from the other bogie as the target domain. Cross-device migration fault diagnosis is performed using a convolutional neural network based on minimizing Shannon entropy and information enhancement. Stable bearing fault data at low speed and generated bearing data under stable working conditions are used as the source domains, and stable bearing fault data at high speed is used as the target domain for domain generalization. The source domain data is trained using the domain generalization method, and fault diagnosis is performed on the target domain data. The accuracy of fault diagnosis is calculated.

[0009] Preferably, the traditional convolutional neural network includes 3 convolutional layers and 1 fully connected layer, and is connected to the traditional convolutional neural network after the feature enhancement layer to perform feature extraction and classification diagnosis on the signal.

[0010] In a second aspect, an embodiment of the present invention provides a bearing fault diagnosis device, comprising: Data module, to obtain the original fault signal of the bearing; The optimization module designs a convolutional neural network that includes Shannon entropy calculation and bearing signal feature enhancement, constructs a filter convolution layer, and optimizes the parameters of the convolutional neural network by minimizing the Shannon entropy of the signal after passing through the filter layer; Enhancement module, calculates the bearing fault frequency, builds a feature enhancement layer, and enhances the bearing fault features; The diagnosis module builds a traditional convolutional layer as a classifier network for fault diagnosis, outputs the fault diagnosis results, and performs migration diagnosis across device domains.

[0011] In a third aspect, an electronic device is provided. The electronic device includes a processor and a memory, wherein the memory stores computer instructions, and when the computer instructions are executed by the processor, the electronic device performs the actions according to the method of the first aspect or any embodiment thereof.

[0012] In a fourth aspect, a computing device cluster is provided, which includes at least one computing device, each computing device including a processor and a memory; the processor of the at least one computing device is used to execute instructions stored in the memory of the at least one computing device, so that the computing device cluster performs operations according to the method in the above-mentioned first aspect or any embodiment thereof.

[0013] In a fifth aspect, a computer-readable storage medium is provided, wherein computer-executable instructions are stored on the computer-readable storage medium, and when the computer-executable instructions are executed by a processor, the operations according to the method of the first aspect or any embodiment thereof are implemented.

[0014] In a sixth aspect, a computer program or computer program product is provided, which is tangibly stored on a computer-readable medium and includes computer-executable instructions that, when executed, implement the operations of the method according to the first aspect or any of its embodiments.

[0015] Compared with the prior art, the present invention has at least the following beneficial effects: A bearing fault diagnosis method decouples signal preprocessing from feature learning through a phased optimization strategy. Adaptive filtering is achieved by minimizing the Shannon entropy criterion, and a dual-path feature optimization architecture is formed in conjunction with a fault frequency feature enhancement mechanism. This step effectively addresses the feature confusion problem in traditional end-to-end models. Signal preprocessing guided by prior information significantly improves the sparsity of fault features, enabling subsequent deep networks to more accurately capture the essential characteristics of fault modes.

[0016] Furthermore, the Shannon entropy optimization module integrates the Shannon entropy metric into the backpropagation process by constructing a differentiable entropy calculation layer, establishing an optimization objective function based on information theory principles. By minimizing this loss function, the convolution kernel is automatically adjusted to the optimal state that minimizes the information entropy of the output signal. This is equivalent to performing blind source separation in parameter space, effectively suppressing noise interference and enhancing the detectability of periodic impulse components.

[0017] Furthermore, the feature enhancement layer constructs a learnable resonator with frequency-selective properties to establish a mapping between the bearing's physical parameters and the network structure. This design embeds domain knowledge into the underlying network, enabling targeted enhancement of fault-sensitive frequency bands and addressing the lack of physical interpretability in traditional data-driven approaches.

[0018] Furthermore, the transfer diagnosis framework constructs a dual-domain adversarial training mechanism, which realizes domain-invariant feature extraction through maximum mean divergence (MMD) or domain adversarial neural network (DANN), so that the model can maintain stable diagnostic performance even when the target domain data distribution is unknown.

[0019] It can be understood that the beneficial effects of the second to sixth aspects mentioned above can be found in the relevant description of the first aspect mentioned above, and will not be repeated here.

[0020] In summary, this invention effectively suppresses noise interference through an adaptive filtering mechanism that minimizes Shannon entropy. It significantly improves feature separability by combining a physical enhancement layer for fault characteristic frequencies, and employs a phased optimization strategy to address feature coupling. While achieving high-precision fault diagnosis, it also constructs a cross-device migration diagnostic framework, ensuring model stability under time-varying conditions. This provides reliable technical support for the intelligent operation and maintenance of high-speed train bearings.

[0021] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0023] Figure 1 A flow chart of a bearing fault diagnosis method based on minimizing Shannon entropy filtering and information enhancement is shown; Figure 2 A signal slice schematic diagram of a bearing fault diagnosis method based on minimized Shannon entropy filtering and information enhancement in an embodiment is shown; Figure 3 A schematic diagram showing the filtering and signal enhancement effects in the embodiment is shown; Figure 4 A schematic block diagram illustrating an example device that may be used to implement embodiments of the present disclosure is shown; Figure 5 A schematic block diagram illustrating an example computing device cluster that may be used to implement embodiments of the present disclosure is shown; Figure 6A schematic block diagram illustrating an example implementation of a cluster of computing devices that may be used to implement embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0024] The following describes embodiments of the present application in more detail with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as limited to the embodiments described herein. Instead, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are for illustrative purposes only and are not intended to limit the scope of protection of the present application.

[0025] In the description of the present invention, it is to be understood that the terms “include” and “comprise” indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or collections thereof.

[0026] It should also be understood that the terms used in the present specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0027] It should be further understood that the term "and / or" as used in the present specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in the present invention generally indicates that the associated objects are in an "or" relationship.

[0028] It should be understood that although the terms "first," "second," and "third" may be used to describe preset ranges in embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are merely used to distinguish one preset range from another. For example, without departing from the scope of embodiments of the present invention, the first preset range may also be referred to as the second preset range, and similarly, the second preset range may also be referred to as the first preset range.

[0029] The word "if," as used herein, may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to the determination" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)," depending on the context.

[0030] The accompanying drawings illustrate various schematic diagrams of structures according to embodiments disclosed herein. These figures are not drawn to scale; for clarity, some details are exaggerated and some details may be omitted. The shapes of the various regions and layers shown in the figures, as well as their relative sizes and positional relationships, are merely exemplary and may deviate in practice due to manufacturing tolerances or technical limitations. Those skilled in the art may design regions / layers with different shapes, sizes, and relative positions as needed.

[0031] This paper provides a bearing fault diagnosis method that achieves high-accuracy fault diagnosis under time-varying operating conditions and exhibits good generalization capabilities. By combining the latest information theory and deep learning techniques, the method optimizes the fault diagnosis model by minimizing Shannon entropy under limited labeled data. Information enhancement techniques are also used to improve the model's ability to identify fault signatures, enabling effective fault diagnosis for diverse equipment and operating conditions.

[0032] Example 1 See also Figure 1 The present invention provides a bearing fault diagnosis method, comprising the following steps: S1. Obtain bearing fault information; Collect fault vibration signals of high-speed train bearings and organize the slices.

[0033] The signal is sliced into 4096 sampling points with a step size of 1024. Overlap sampling is used for data enhancement to increase the diversity of data samples, thereby improving the robustness and generalization ability of the model and ensuring that the collected fault signal information can be fully utilized during the training process.

[0034] S2. Design a convolutional neural network that includes Shannon entropy calculation and bearing signal feature enhancement, and optimize the filter convolution layer by minimizing Shannon entropy; S201, constraining convolution kernel parameters to design a filter convolution layer, and implementing filtering processing on the original fault signal using a convolutional neural network; First, a filtering convolution layer is constructed. The core of the convolution layer lies in its convolution kernel. The convolution layer calculates the weighted sum of the local area by sliding convolution on the input data. By constraining the convolution kernel parameters, the signal can be filtered. The formula for the filtering convolution kernel coefficient is as follows:

[0035] Among them, the key parameters of the filter convolution kernel include shape factors σ and k. By optimizing the shape factors σ and k of the convolution kernel, the shape of the convolution kernel is changed to achieve filtering of different frequencies.

[0036] S202, by minimizing the Shannon entropy of the signal after passing through the filter layer, optimizing the parameters of the filter convolution layer, and improving the model's ability to distinguish signals; Input the vibration signal data into the filter convolution layer and calculate the Shannon entropy of the output signal. The Shannon entropy calculation formula is as follows:

[0037] in, It's a signal Middle Status The probability of occurrence, Shannon entropy measures the average uncertainty and information content of the signal.

[0038] By minimizing the Shannon entropy and optimizing the filter convolution kernel parameters, the Shannon entropy of the output signal after the filter layer is minimized as much as possible, thereby improving the signal-to-noise ratio and facilitating the feature extraction of the network.

[0039] S203, inputting the signal of the filter convolution layer through minimizing Shannon entropy optimization into the subsequent feature extraction and classification network, and the Shannon entropy minimization optimization process is performed separately from the subsequent model feature extraction process.

[0040] S3. Calculate the bearing fault frequency, construct a feature enhancement layer, enhance the signal fault frequency position information, and perform feature enhancement on the bearing fault features; First, the fault characteristic frequency of the bearing is calculated. The fault characteristic frequencies of the inner ring, outer ring, rolling element and roller can be calculated through the model shape parameters. The convolution kernel parameters are constrained to construct a feature enhancement layer, so that this layer can enhance the signals at each fault characteristic frequency of the incoming data.

[0041] S301. Calculate the characteristic frequency of the bearing based on the bearing size and shape data. The calculation formula for the bearing fault frequency is as follows: Inner race fault frequency

[0042] Outer race fault frequency

[0043] Roller failure frequency

[0044] Cage failure frequency

[0045] S302. Design a feature enhancement layer to enhance the signal features using the frequency characteristics of the bearing rotation frequency, inner race fault frequency (BPFI), outer race fault frequency (BPFO), and roller fault frequency (BSF). This layer enhances the signal strength at the fault feature frequency. If a certain type of fault occurs in the signal, the fault feature of the signal will be more obvious after passing through the signal enhancement layer, which will help the network extract the signal features.

[0046] The convolution kernel coefficient formula of the feature enhancement layer is similar to that of the filter convolution layer:

[0047] However, compared with the filtering convolution layer, the feature enhancement layer constrains the center frequency and does not change the signal strength of other frequency bands. It only enhances the signal at the fault feature.

[0048] S4. Output the fault diagnosis result; After the feature enhancement layer, a traditional convolutional neural network is connected to perform feature extraction and classification diagnosis on the signal.

[0049] A simple traditional convolutional neural network is constructed, including three convolutional layers and one fully connected layer. The traditional convolutional neural network is connected after the feature enhancement layer to perform feature extraction and classification diagnosis on the signal.

[0050] S5. Perform migration diagnosis across devices.

[0051] Cross-device fault diagnosis using a model based on minimizing Shannon entropy and information enhancement.

[0052] S501, obtaining vibration data of two different bogies, using the data of one bogie as a source domain and the data of the other bogie as a target domain; The above model based on minimizing Shannon entropy and information enhancement is used to perform cross-device migration fault diagnosis. The stable bearing fault data at low speed and the generated stable working condition bearing data are used as the source domain, and the stable bearing fault data at high speed is used as the target domain for domain generalization; S502: Use the domain generalization methods mmd, lmmd, dann, and bnm to train the model on the source domain data, perform fault diagnosis on the target domain data, and calculate the accuracy of the fault diagnosis.

[0053] Those skilled in the art will appreciate that various aspects of the present invention may be implemented as systems, methods, or program products. Accordingly, various aspects of the present invention may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, microcode, etc.), or in a combination of hardware and software, collectively referred to herein as "circuits," "modules," or "platforms."

[0054] Example 2 The present invention provides a bearing fault diagnosis device, which can be used to implement the above-mentioned bearing fault diagnosis method. Specifically, the bearing fault diagnosis device includes a data module, an optimization module, an enhancement module and a diagnosis module.

[0055] Among them, the data module obtains the original fault signal of the bearing; The optimization module designs a convolutional neural network that includes Shannon entropy calculation and bearing signal feature enhancement, constructs a filter convolution layer, and optimizes the parameters of the convolutional neural network by minimizing the Shannon entropy of the signal after passing through the filter layer; Enhancement module, calculates the bearing fault frequency, builds a feature enhancement layer, and enhances the bearing fault features; The diagnosis module builds a traditional convolutional layer as a classifier network for fault diagnosis, outputs the fault diagnosis results, and performs migration diagnosis across device domains.

[0056] Example 3 The embodiment of the present disclosure further provides a computing device 900. Figure 4 As shown, computing device 900 includes a bus 902, a processor 904, a memory 906, and a communication interface 908. Processor 904, memory 906, and communication interface 908 communicate with each other via bus 902. Computing device 900 can be a server or a terminal device. It should be understood that this application does not limit the number of processors and memories in computing device 900.

[0057] The bus 902 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 The bus 904 may include a path for transmitting information between various components of the computing device 900 (eg, memory 906, processor 904, communication interface 908).

[0058] The processor 904 may include any one or more processors such as a central processing unit (CPU), a graphics processing unit (GPU), a tensor processing unit (TPU), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a microprocessor (MP), or a digital signal processor (DSP).

[0059] The memory 906 may include volatile memory, such as random access memory (RAM). The processor 904 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0060] The memory 906 stores executable program codes, and the processor 904 executes the executable program codes to implement, for example, a bearing fault diagnosis method. That is, the memory 906 may store instructions for the methods and functions related to the computing device 110 in any of the above embodiments.

[0061] The communication interface 908 uses a transceiver module such as, but not limited to, a network interface card or a transceiver to implement communication between the computing device 900 and other devices or a communication network.

[0062] Example 4 Embodiments of the present disclosure also provide a computing device cluster 1000. The computing device cluster includes at least one computing device. The computing device can be a server, such as a central server, an edge server, or a local server in a local data center. In some embodiments, the computing device can also be a terminal device such as a desktop computer, a laptop computer, or a smartphone.

[0063] like Figure 5As shown, the computing device cluster includes at least one computing device 900. The memory 906 in one or more computing devices 900 in the computing device cluster may store the same instructions for executing the methods and functions related to the computing device 110 in any of the above embodiments.

[0064] In some possible implementations, the memory 906 of one or more computing devices 900 in the computing device cluster may also store partial instructions for executing the methods and functions related to the computing device 110 in any of the above embodiments. In other words, the combination of one or more computing devices 900 can jointly execute instructions for executing the methods and functions of the computing device 110.

[0065] It should be noted that the memories 906 in different computing devices 900 in the computing device cluster may store different instructions, respectively used to execute part of the functions of the apparatus 800. That is, the instructions stored in the memories 906 in different computing devices 900 may implement the bearing fault diagnosis method.

[0066] In some possible implementations, one or more computing devices in a computing device cluster may be connected via a network, which may be a wide area network or a local area network. Figure 6 A possible implementation 1100 is shown. Figure 6 As shown, two computing devices 900A and 900B are connected via a network 1110. Specifically, each computing device is connected to the network via a communication interface within the computing device. In one possible implementation, for example, the memory 906 within computing device 900A stores execution instructions. Simultaneously, the memory 906 within computing device 900B stores instructions for executing the functions of adjustment module 830.

[0067] Figure 6 The connection method between the computing device clusters shown may be based on the consideration that the method provided in this application regarding the computing device 110 may require storage of a large amount of data and intensive computing, and therefore it is considered that the functions implemented by the adjustment module 830 may be executed by the computing device 900B.

[0068] It should be understood that Figure 6 The functions of the computing device 900A shown in the figure may also be performed by multiple computing devices 900. Similarly, the functions of the computing device 900B may also be performed by multiple computing devices 900. The embodiments of the present disclosure also provide a computer program product including instructions, which, when executed on a computer, enables the computer to perform the methods and functions involving the computing device 110 in any of the above embodiments.

[0069] Example 5 An embodiment of the present disclosure further provides a computer-readable storage medium having computer instructions stored thereon. When a processor executes the instructions, the processor executes the methods and functions involving the computing device 110 in any of the above embodiments.

[0070] In general, various embodiments of the present disclosure may be implemented in hardware or dedicated circuitry, software, logic, or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software, which may be executed by a controller, microprocessor, or other computing device. Although various aspects of the embodiments of the present disclosure are shown and described as block diagrams, flow charts, or using some other pictorial representation, it should be understood that the blocks, devices, systems, techniques, or methods described herein may be implemented as, by way of non-limiting example, hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or a controller or other computing device, or some combination thereof.

[0071] The present disclosure also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which are executed in a device on a real or virtual processor of a target to perform the process / method as described above with reference to the accompanying drawings. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided between program modules as needed. The machine-executable instructions for the program modules can be executed in local or distributed devices. In distributed devices, program modules can be located in local and remote storage media.

[0072] The computer program code for implementing the disclosed method can be written in one or more programming languages. These computer program codes can be provided to the processor of a general-purpose computer, a special-purpose computer or other programmable data processing device so that the program code, when executed by the computer or other programmable data processing device, causes the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code can be executed entirely on a computer, partially on a computer, as an independent software package, partially on a computer and partially on a remote computer or entirely on a remote computer or server.

[0073] In the context of the present disclosure, computer program code or related data may be carried by any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals may include electrical, optical, radio, acoustic, or other forms of propagated signals, such as carrier waves, infrared signals, and the like.

[0074] A computer-readable medium may be any tangible medium containing or storing a program for or relating to an instruction execution system, apparatus, or device, or a data storage device such as a data center containing one or more available media. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination thereof. More detailed examples of computer-readable storage media include an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0075] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0076] The CWRU Western Reserve University bearing dataset was used for validation, with four load conditions: 0hp, 1hp, 2hp, and 3hp. Cross-domain testing was conducted using the 1hp and 2hp conditions as the source domains and the 3hp condition as the target domain. Within the -4dB to 20dB signal-to-noise ratio range, the model using filtered convolutional layers performed better than traditional CNN models in various noise environments. Under -4dB conditions, the proposed method achieved a 7.3% improvement in accuracy compared to the baseline model, and the accuracy degradation with decreasing signal-to-noise ratio was less than that of the baseline model, validating the proposed method's ability to resist noise interference.

[0077] Comparing the performance of the proposed method on six typical migration models (MMD, DANN, DSN, ADDA, DeepCORAL, and CDAN), the average migration accuracy of the proposed method under cross-speed conditions reached 91.4%, an increase of 6.2 percentage points compared with the baseline model, verifying the robustness of the proposed feature enhancement mechanism to domain shift.

[0078] The present invention first obtains bearing fault information; designs a convolutional neural network including Shannon entropy calculation and bearing signal feature enhancement; optimizes the filtering convolution layer by minimizing Shannon entropy; enhances the bearing fault features; outputs the fault diagnosis results; performs migration diagnosis across equipment fields to solve the current high-speed rail bogie bearing fault model, the low cross-bogie accuracy and generalization problems; solves the problem of low field generalization accuracy of bearing fault diagnosis under time-varying working conditions, thereby improving the accuracy and reliability of bearing fault diagnosis.

[0079] In summary, the present invention provides a bearing fault diagnosis method, equipment, cluster, medium, program product, and device. By combining the information entropy optimization criterion with the physical feature enhancement mechanism, a deep diagnostic framework with clear physical meaning is constructed. The innovation is reflected in: (1) A differentiable Shannon entropy calculation module is proposed to realize adaptive filtering; (2) Establishing an explicit association between bearing dynamic characteristics and network parameters; (3) Design a two-stage optimization strategy to solve the feature coupling problem.

[0080] Experiments have shown that the method of the present invention has significant advantages in strong noise and domain offset scenarios, and provides a new technical path for intelligent diagnosis under complex working conditions.

[0081] In addition, although the operations of the method of the present disclosure are described in a particular order in the accompanying drawings, this does not require or imply that these operations must be performed in this particular order, or that all the operations shown must be performed to achieve the desired result. On the contrary, the steps depicted in the flowchart can change the order of execution. Additionally or alternatively, certain steps can be omitted, multiple steps can be combined into one step, and / or one step can be decomposed into multiple steps. It should also be noted that the features and functions of two or more devices according to the present disclosure can be embodied in one device. Conversely, the features and functions of a device described above can be further divided into being embodied by multiple devices.

[0082] While various implementations of the present disclosure have been described above, the foregoing description is intended to be illustrative, non-exhaustive, and non-limiting to the disclosed implementations. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described implementations. The terminology used herein is selected to facilitate explanation of the principles, practical applications, or improvements to existing technologies, or to enable others skilled in the art to understand the various implementations disclosed herein.

Claims

1. A bearing fault diagnosis method, characterized in that: The following steps are involved: Obtain the original fault signal of the bearing; Design a convolutional neural network that includes Shannon entropy calculation and bearing signal feature enhancement, construct a filtering convolution layer, and optimize the parameters of the convolutional neural network by minimizing the Shannon entropy of the signal after passing through the filtering layer; Calculate the bearing fault frequency and build a feature enhancement layer to enhance the bearing fault features; Construct traditional convolutional layers as classifier networks for fault diagnosis, output fault diagnosis results, and perform migration diagnosis across device domains.

2. The bearing fault diagnosis method according to claim 1, characterized in that: A convolutional neural network is designed that includes Shannon entropy calculation and bearing signal feature enhancement. A filter convolution layer is constructed, and the parameters of the convolutional neural network are optimized by minimizing the Shannon entropy of the signal after the filter layer. Specifically: Construct a filtering convolution layer, which calculates the weighted sum of the local area by sliding convolution on the input data and performs filtering on the signal by constraining the convolution kernel parameters; Input the vibration signal data into the filter convolution layer and calculate the Shannon entropy of the output signal; The signal of the filtered convolution layer optimized by minimizing Shannon entropy is input into the subsequent feature extraction and classification network. The Shannon entropy minimization optimization process is carried out separately from the subsequent model feature extraction process.

3. The bearing fault diagnosis method according to claim 1, characterized in that: Construct a feature enhancement layer to enhance the bearing fault features, specifically: The characteristic frequency of the bearing is calculated based on the bearing size and shape data; a feature enhancement layer is designed to enhance the signal characteristics using the frequency characteristics of the bearing rotation frequency, inner ring fault frequency, outer ring fault frequency, and roller fault frequency.

4. The bearing fault diagnosis method according to claim 1, characterized in that: Construct a traditional convolutional layer as a classifier network for fault diagnosis, output the fault diagnosis results, and perform migration diagnosis across devices. Specifically: Vibration data from two different bogies are obtained, with the data from one bogie as the source domain and the data from the other bogie as the target domain. Cross-device migration fault diagnosis is performed using a convolutional neural network based on minimizing Shannon entropy and information enhancement. Stable bearing fault data at low speed and generated bearing data under stable working conditions are used as the source domains, and stable bearing fault data at high speed is used as the target domain for domain generalization. The source domain data is trained using the domain generalization method, and fault diagnosis is performed on the target domain data. The accuracy of fault diagnosis is calculated.

5. The bearing fault diagnosis method according to claim 4, characterized in that: The traditional convolutional neural network, including three convolutional layers and one fully connected layer, is connected to the feature enhancement layer to perform feature extraction and classification diagnosis on the signal.

6. A bearing fault diagnosis device, characterized in that: include: Data module, to obtain the original fault signal of the bearing; The optimization module designs a convolutional neural network that includes Shannon entropy calculation and bearing signal feature enhancement, constructs a filter convolution layer, and optimizes the parameters of the convolutional neural network by minimizing the Shannon entropy of the signal after passing through the filter layer; Enhancement module, calculates the bearing fault frequency, builds a feature enhancement layer, and enhances the bearing fault features; The diagnosis module builds a traditional convolutional layer as a classifier network for fault diagnosis, outputs the fault diagnosis results, and performs migration diagnosis across device domains.

7. An electronic device, characterized in that: The electronic device comprises a processor and a memory, wherein computer instructions are stored in the memory. When the computer instructions are executed by the processor, the electronic device executes the method according to any one of claims 1 to 5.

8. A computing device cluster, characterized in that: comprising at least one computing device, each computing device including a processor and a memory; The processor of the at least one computing device is configured to execute instructions stored in a memory of the at least one computing device, so that the computing device cluster performs the method according to any one of claims 1 to 5.

9. A computer program product, characterized in that The computer program product contains computer-executable instructions, which implement the method according to any one of claims 1 to 5 when executed.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which implement the method according to any one of claims 1 to 5 when executed by a processor.

Images