Intelligent Fault Diagnosis Platform Based on Virtual Test Stands and End-to-End Cloud Link

Abstract

This application relates to an end-to-end intelligent fault diagnosis platform based on a virtual test bench. The platform includes: a virtual test bench system for dynamic and fault type modeling of the system or component under test, simulating its response characteristics under set boundary conditions and fault states, and converting the response characteristics into electrical signals; an edge computing system for preprocessing and feature extraction of the acquired electrical signals; and a cloud-based intelligent fault early warning system for users to perform fault diagnosis, reliability analysis, or remaining lifetime estimation based on the features extracted by the edge computing system. This platform, based on a virtual test bench model to verify the end-to-end technology, significantly reduces experimental costs and improves the efficiency of early testing and deployment of diagnostic algorithms. Its modular structure allows for modification and replacement of algorithms and models within modules as needed, and the communication methods between modules can also be configured as required, exhibiting excellent scalability and adaptability.

Description

Technical Field

[0001] This application relates to the field of fault diagnosis technology, and in particular to an intelligent fault diagnosis platform based on a virtual test bench across the entire end-to-edge-to-cloud link. Background Technology

[0002] With the widespread application of artificial intelligence algorithms, such as machine learning and deep learning, in equipment fault diagnosis and early warning, the size of models and the complexity of algorithms are increasing to ensure the accuracy of fault diagnosis, while the resources and computing power of embedded systems are limited. Therefore, deploying intelligent diagnostic algorithms on cloud servers and realizing communication between the device under test and the cloud server through edge computing systems to build an intelligent diagnostic system based on end-edge-cloud collaboration has become a research hotspot and technological trend in the field of PHM (Prognostics and Health Management).

[0003] Extensive and in-depth research has been conducted in academia and industry on end-to-end intelligent fault diagnosis platforms. Document CN116106005A discloses an end-to-end-cloud collaborative fault diagnosis framework. By deploying TinyML models on end nodes, it effectively utilizes the computing power of end devices, reduces the computational and communication costs after model deployment, and resolves the contradiction between fault diagnosis accuracy and latency. Document CN115186883A discloses an industrial equipment health status monitoring system and method based on edge-cloud collaborative computing, providing an implementation process from edge data reading to cloud fault diagnosis. Document CN112101532A discloses an adaptive multi-model driven equipment fault diagnosis method based on edge-cloud collaboration. It divides the deep learning-based fault diagnosis model into layers between the edge and cloud, proposing a cross-condition diagnosis method based on edge-cloud collaboration. By training a general condition model in the cloud and distributing it to the edge, the edge then diagnoses its personalized condition data, effectively reducing the latency problem of edge-cloud collaborative fault diagnosis. The literature (announcement number CN113567159A) proposes an edge-cloud collaborative method for the condition monitoring and fault diagnosis of scraper conveyors. By building a distributed neural network model DDNN composed of an edge-side network model and a cloud-side network model, it effectively solves the problem that neural network models cannot be directly used for IoT devices.

[0004] While achieving some results, the aforementioned inventions all have certain limitations. For example, most of the publicly available research focuses on edge data processing and cloud-based intelligent diagnostic algorithms, while the "edge" in the "edge-cloud" framework is rarely addressed. Existing research generally assumes that the "edge" is the object under test in actual engineering projects. However, in most application scenarios, building an "edge-cloud" system based on an actual object under test or test bench is not only expensive and time-consuming to debug, but also hinders the rapid verification of the diagnostic effectiveness of intelligent algorithms in the cloud or edge-cloud collaboration. Furthermore, the operating conditions and boundary conditions of actual equipment are often limited to a few typical conditions for safety reasons, which is not conducive to fully testing the performance of diagnostic algorithms under the "edge-cloud" framework. Summary of the Invention

[0005] Therefore, it is necessary to provide an intelligent fault diagnosis platform based on a virtual test bench that covers the entire end-to-end-to-cloud link to address the aforementioned technical issues.

[0006] An intelligent fault diagnosis platform based on a virtual test bench and spanning the entire edge-cloud chain includes: a virtual test bench system, an edge computing system, and a cloud-based intelligent fault early warning system.

[0007] A virtual test bench system is used to perform dynamic modeling and fault type modeling of the system or component under test, and to simulate the response characteristics of the system or component under test under set boundary conditions and fault states, and convert the response characteristics into electrical signals.

[0008] The edge computing system is used to preprocess and extract features from the collected electrical signals and transmit the extracted features to the cloud-based intelligent fault early warning system. It is also used to receive the results of fault diagnosis, reliability analysis, or remaining lifetime estimation transmitted from the cloud-based intelligent fault early warning system, so that users can analyze the diagnostic results of the tested system or component at the edge and improve the data preprocessing and feature extraction methods and parameter configurations at the edge based on the diagnostic results, thereby further optimizing the diagnostic results in the cloud.

[0009] The cloud-based intelligent fault early warning system is used to perform fault diagnosis, reliability analysis, or remaining lifetime estimation on the tested system or component based on the features extracted by the edge computing system, and transmits the results of fault diagnosis, reliability analysis, or remaining lifetime estimation to the edge computing system.

[0010] In one embodiment, the virtual test bench system includes: a dynamic model, a fault model, and a hardware interface.

[0011] A dynamic model is used to establish the kinematic and dynamic equations of the measured system or component under various degrees of freedom based on Newton's laws; the kinematic and dynamic equations are used to characterize the response characteristics of the system under different boundary conditions.

[0012] Fault models are used to model wear and aging faults in the system or component under test.

[0013] The hardware interface is used to convert the response characteristics of the system or component under test, which is simulated using dynamic and fault models, into electrical signals under set and fault conditions, and transmit the electrical signals to the edge computing system via the TCP / IP protocol.

[0014] In one embodiment, the edge computing system includes a hardware module consisting of an embedded system or edge computing components and a signal processing algorithm module running on the hardware module; the signal processing algorithm module includes a signal preprocessing submodule, a feature extraction submodule, and a feature data / file transfer submodule.

[0015] The signal preprocessing submodule is used to preprocess the acquired electrical signals.

[0016] The feature extraction submodule is used to extract fault features from the time-frequency domain of the preprocessed signal.

[0017] The feature data / file transfer submodule is used to transmit the fault feature data extracted by the feature extraction module to the cloud-based intelligent fault early warning system.

[0018] In one embodiment, the cloud-based intelligent fault early warning system includes a fault diagnosis hardware module composed of a cloud computing platform or a high-performance server and a fault diagnosis algorithm module running on the fault diagnosis hardware module; the fault diagnosis algorithm module includes: a fault diagnosis and early warning submodule, a visualization processing submodule, and a fault data / file transfer submodule.

[0019] The fault diagnosis and early warning submodule is used to reason and analyze the fault feature data sent by the edge computing system using machine learning methods, deep learning methods, or a combination of machine learning and deep learning methods, so as to realize fault classification, reliability assessment and remaining life estimation of the tested system or component.

[0020] The visualization processing submodule is used to package the process results and final results of the fault diagnosis and early warning submodule into visual charts or files.

[0021] The fault data / file transfer submodule is used to transfer visualized charts or files back to the edge computing system via the SFTP protocol.

[0022] The aforementioned end-to-edge-cloud intelligent fault diagnosis platform based on a virtual test bench comprises a virtual test bench system, an edge computing system, and a cloud-based intelligent fault early warning system. The virtual test bench system performs dynamic and fault type modeling on the system or component under test (SUT), and simulates its response characteristics under set boundary conditions and fault states, converting these response characteristics into electrical signals. The edge computing system preprocesses and extracts features from the acquired electrical signals, transmitting the extracted features to the cloud-based intelligent fault early warning system. The cloud-based intelligent fault early warning system performs fault diagnosis, reliability analysis, or remaining lifetime estimation on the SUT based on the features extracted by the edge computing system. This platform can validate the end-to-edge-cloud technology based on a virtual test bench model, significantly reducing experimental costs and improving the efficiency of early testing and deployment of diagnostic algorithms. The platform adopts a modular structure, allowing algorithms and models within modules to be modified and replaced according to actual needs, and the communication methods between modules can also be configured as required, exhibiting excellent scalability and adaptability. Attached Figure Description

[0023] Figure 1 This is a block diagram of an intelligent fault diagnosis platform based on a virtual test bench across the entire end-to-end-to-cloud chain, as shown in one embodiment.

[0024] Figure 2 This is a schematic diagram of the hardware interface of a virtual test bench in one embodiment;

[0025] Figure 3 The following is a simulation result of a fault on a virtual test bench in one embodiment, where (a)-(d) represent the bearing acceleration response under normal conditions, inner ring fault, outer ring fault, and ball fault, respectively.

[0026] Figure 4 This is a fault classification result diagram based on the "end-edge-cloud" platform in another embodiment;

[0027] Figure 5 This is a graph showing the remaining lifetime estimation results based on the "end-edge-cloud" platform in one embodiment; Detailed Implementation

[0028] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0029] In one embodiment, such as Figure 1As shown, an intelligent fault diagnosis platform based on a virtual test bench and covering the entire end-to-end-to-cloud chain is provided. The platform includes: a virtual test bench system 10, an edge computing system 20, and a cloud-based intelligent fault early warning system 30.

[0030] The virtual test bench system 10 is used to perform dynamic modeling and fault type modeling of the system or component under test, and to simulate the response characteristics of the system or component under test under set boundary conditions and fault states, and to convert the response characteristics into electrical signals.

[0031] Specifically, in order to reduce the deployment and testing costs of the "device-edge-cloud" end-to-end platform and improve the testing efficiency of edge-cloud collaborative algorithms, a virtual test bench model for intelligent diagnosis of the "device-edge-cloud" end-to-end is proposed.

[0032] The virtual test bench model consists of three parts: a dynamic model, a fault model, and a hardware interface. The dynamic model is responsible for simulating the response characteristics of the system under normal conditions, the fault model is used to model the common faults and aging characteristics of the system, and the hardware interface is responsible for converting the output generated by the simulation model into electrical signals for real-time acquisition by the edge computing system.

[0033] Virtual test benches are used to simulate the sensor outputs of a system under normal and fault conditions.

[0034] The edge computing system 20 is used to preprocess and extract features from the collected electrical signals and transmit the extracted features to the cloud-based intelligent fault early warning system 30. It is also used to receive the results of fault diagnosis, reliability analysis or remaining life estimation transmitted by the cloud-based intelligent fault early warning system 30, so that users can analyze the diagnostic results of the tested system or component at the edge and improve the data preprocessing and feature extraction methods and parameter configurations at the edge based on the diagnostic results, and further optimize the diagnostic results in the cloud.

[0035] The cloud-based intelligent fault early warning system 30 is used to perform fault diagnosis, reliability analysis, or remaining lifetime estimation on the system or component under test based on the features extracted by the edge computing system 20, and transmits the results of fault diagnosis, reliability analysis, or remaining lifetime estimation to the edge computing system 20.

[0036] The aforementioned end-to-edge-cloud intelligent fault diagnosis platform based on a virtual test bench comprises a virtual test bench system, an edge computing system, and a cloud-based intelligent fault early warning system. The virtual test bench system performs dynamic and fault type modeling on the system or component under test (SUT), and simulates its response characteristics under set boundary conditions and fault states, converting these response characteristics into electrical signals. The edge computing system preprocesses and extracts features from the acquired electrical signals, transmitting the extracted features to the cloud-based intelligent fault early warning system. The cloud-based intelligent fault early warning system performs fault diagnosis, reliability analysis, or remaining lifetime estimation on the SUT based on the features extracted by the edge computing system. This platform can validate the end-to-edge-cloud end-to-end technology based on the virtual test bench model, significantly reducing experimental costs and improving the efficiency of early testing and deployment of diagnostic algorithms. The platform adopts a modular structure, allowing algorithms and models within modules to be modified and replaced according to actual needs, and the communication methods between modules can also be configured as required, exhibiting excellent scalability and adaptability.

[0037] In one embodiment, the virtual test bench system includes: a dynamic model, a fault model, and a hardware interface; the dynamic model is used to establish the kinematic and dynamic equations of the tested system or component under various degrees of freedom according to Newton's laws; the kinematic and dynamic equations are used to characterize the response characteristics of the system under different boundary conditions; the fault model is used to model wear faults and aging faults of the tested system or component; the hardware interface is used to convert the response characteristics of the tested system or component simulated by the dynamic model and fault model under set conditions and fault conditions into electrical signals, and transmit the electrical signals to the edge computing system via the TCP / IP protocol.

[0038] Specifically, virtual test bench setup and testing includes system dynamics modeling, system fault modeling, hardware interface design, virtual test bench model integration and testing, etc. To facilitate the explanation of the implementation plan, a bearing is used as an example to detail the implementation steps. Other systems or components under test can also refer to this method and process for implementation.

[0039] (1) System dynamics modeling

[0040] A system dynamics model consists of the kinematic and dynamic equations of a system under each degree of freedom, used to characterize the system's response characteristics under different boundary conditions. Generally, the kinematic and dynamic equations of a system under each degree of freedom are established based on Newton's laws. This embodiment uses the 5-degree-of-freedom dynamic equations of a bearing as an example to illustrate system dynamics modeling. The 5 degrees of freedom include: the horizontal and vertical degrees of freedom of the inner and outer rings of the bearing, and the vertical degree of freedom of the shock absorber. Based on Newton's second law, the following equations can be established:

[0041]

[0042] Where: m s m p m r These are the masses of the bearing inner ring, bearing outer ring, and shock absorber, respectively; k s k p k r These are the stiffnesses of the bearing inner ring, bearing outer ring, and shock absorber, respectively; c s c p c r These are the damping components for the inner ring of the bearing, the outer ring of the bearing, and the shock absorber, respectively; x s x p y s ,y p These represent the displacements of the inner and outer rings of the bearing in the x and y directions, respectively. b F represents the displacement of the shock absorber in the y-direction. x and F y External forces in the x and y directions, respectively; f x f y These represent the contact forces between the ball and the raceway in the x and y directions, respectively.

[0043] The contact force f between the ball and the raceway in the x and y directions. x f y The contact force f between the ball and the raceway in the x and y directions can be determined using Herz contact theory. x f y for:

[0044]

[0045] Where, k b δ represents the stiffness of the bearing balls. j For the deformation of the j-th ball, φ j Let γ be the angular position of the j-th ball. j For switching functions, such as δ j >0, γ j =1, otherwise, γ j =0.

[0046] (2) System Fault Modeling

[0047] Common mechanical system failure types include wear failure and aging failure. The principles of wear failure modeling and aging failure modeling are introduced below.

[0048] 1) Wear Fault Modeling

[0049] From formulas (1) and (2) above, it can be seen that the Herz contact force between the ball and the raceway is mainly determined by the contact deformation, and the deformation of each ball can be calculated from the relative displacement of the inner and outer raceways at that position and the bearing clearance. The deformation of the j-th ball is:

[0050] δ j =(x s -x p cosφ j +9y s -y p sinφ j -c (3)

[0051] When a bearing fails, the location, size, and number of faults further affect the deformation between the balls and raceways. Therefore, the ball deformation of a failed bearing can be modeled as follows:

[0052] δ′ j =(x s -x p cosφ j +(y s -y p sinφ j -c-β j c d (4)

[0053] Where c is the bearing clearance, β j Due to the location of the fault, c d Influenced by the size and shape of the fault. Therefore, further adjustments to β... j and c d Modeling is used to model wear failures.

[0054] 2) Aging Fault Modeling

[0055] For aging failures in mechanical systems, a common modeling method is to use an exponential aging function. The specific method is shown in equation (5):

[0056]

[0057] Where h(t) is the health status index of the bearing, θ and β are coefficients characterizing the aging rate, and ∈ represents normally distributed noise with a mean of 0 and a variance of σ. 2 , This is a constant term. Based on this aging model and combined with collected historical data, a model can be fitted to a specific bearing, thereby achieving modeling of its aging characteristics.

[0058] (3) Hardware Interface

[0059] A virtual test bench system, consisting of a normal dynamics model and an aging fault model, can simulate the system's response characteristics under set boundary conditions and fault states. To transmit the virtual test bench's output signal to the edge computing system, a hardware interface needs to be added to the virtual test bench system model. For ease of explanation, this section uses the mainstream modeling software Matlab / Simulink to develop a virtual test bench model and an edge computing system built on a Raspberry Pi as examples to introduce the design process of the hardware interface.

[0060] like Figure 2 As shown, first, based on the signal type output by the virtual test bench (such as AD, I / O), select the corresponding hardware interface module. For example, the Simulink module for I / O signal acquisition for the Raspberry Pi 4B is GPIO. Second, configure parameters such as the number of channels, sampling frequency, and sampling accuracy according to the signal acquisition requirements of the actual scenario. Similarly, other hardware interfaces for signals that need to be acquired can be configured.

[0061] (4) Virtual bench model integration and testing

[0062] After completing the virtual test bench modeling and hardware interface configuration, each hardware interface module can be connected to the corresponding output signal in the virtual test bench model, thus completing the integration of the virtual test bench model. Then, by setting the corresponding boundary conditions and fault parameters, the integrated virtual test bench model is run, and the signals output by each hardware interface are checked to see if they meet the requirements in terms of type, accuracy, frequency, etc.

[0063] In one embodiment, the edge computing system includes a hardware module consisting of an embedded system or edge computing components and a signal processing algorithm module running on the hardware module; the signal processing algorithm module includes a signal preprocessing submodule, a feature extraction submodule, and a feature data / file transfer submodule.

[0064] The signal preprocessing submodule is used to preprocess the acquired electrical signals.

[0065] The feature extraction submodule is used to extract fault features from the time-frequency domain of the preprocessed signal.

[0066] The feature data / file transfer submodule is used to transmit the fault feature data extracted by the feature extraction module to the cloud-based intelligent fault early warning system.

[0067] Specifically, the construction and testing of the edge computing system includes the design of the signal preprocessing module, the feature extraction module, the data / file transfer module, the integration and testing of the edge computing system modules, etc. The implementation plan for each step will be described in detail below.

[0068] (1) Signal preprocessing module

[0069] The signal preprocessing module is mainly responsible for preprocessing the signals acquired by the edge computing system, specifically including signal filtering and noise reduction, outlier removal, and sample segmentation. For each step, there are various mature methods available in academia and industry. For example, filtering and noise reduction can employ mean filtering, bandpass filtering, wavelet filtering, etc. Users only need to choose a suitable method. This is not the core content of this application and will not be elaborated upon here.

[0070] (2) Feature extraction module

[0071] The feature extraction module is primarily responsible for extracting features from the preprocessed signal in the time-frequency domain to reduce the burden of data transmission between the edge and the cloud server, thereby reducing latency and improving efficiency. Regarding feature extraction methods, many widely used methods exist in the PHM field, such as time-frequency domain metrics like mean, standard deviation, spectral density, and kurtosis, as well as frequency domain fault characteristic frequencies, spectral kurtosis, and wavelet coefficients. Users only need to select one or more suitable methods; this is not the core content of this application and will not be elaborated upon here.

[0072] (3) Data / file transfer module

[0073] To improve the data transmission rate from the edge to the cloud and reduce latency and packet loss, this application sends the extracted features, rather than the raw data collected by the edge system, to the cloud server. The data transmission uses the SFTP protocol. First, the IP address, port number, username, and password of the target cloud platform are configured in the edge computing system to establish a connection between the edge computing system and the cloud platform. Then, the extracted features are sent to the cloud server as files via SFTP. The SFTP protocol provided here can be replaced with other protocols or tools that support edge-cloud communication as needed.

[0074] (4) Edge computing system module integration and testing

[0075] The signal preprocessing module, feature extraction module, and data / file transfer module are integrated, and a main function is defined to call all functional modules. By defining the main function as a periodically executed task, for example, with an execution cycle of 2ms, the entire process from signal processing to edge-cloud data transmission is automated. Finally, by sending fault data from a virtual test bench to the edge computing system, the signal preprocessing, feature extraction, and data / file transfer functions of the edge computing system are tested and verified to ensure that the data received by the cloud server meets expectations.

[0076] In one embodiment, the cloud-based intelligent fault early warning system includes a fault diagnosis hardware module composed of a cloud computing platform or a high-performance server and a fault diagnosis algorithm module running on the fault diagnosis hardware module; the fault diagnosis algorithm module includes: a fault diagnosis and early warning submodule, a visualization processing submodule, and a fault data / file transfer submodule.

[0077] The fault diagnosis and early warning submodule is used to reason and analyze the fault feature data sent by the edge computing system using machine learning methods, deep learning methods, or a combination of machine learning and deep learning methods, so as to realize fault classification, reliability assessment and remaining life estimation of the tested system or component.

[0078] The visualization processing submodule is used to package the process results and final results of the fault diagnosis and early warning submodule into visual charts or files.

[0079] The fault data / file transfer submodule is used to transfer visualized charts or files back to the edge computing system via the SFTP protocol.

[0080] Specifically, the construction and testing of the cloud-based intelligent fault early warning platform includes four steps: fault diagnosis and early warning module design, visualization processing module design, data / file transfer module design, and cloud-based intelligent fault early warning platform integration and testing. The implementation plan for each step will be described in detail below.

[0081] (1) Fault diagnosis and early warning module

[0082] The fault diagnosis and early warning module is the core of the cloud-based intelligent fault early warning platform. It is primarily responsible for further reasoning and analysis of fault characteristic data sent from the edge computing system to the cloud server, enabling tasks such as fault classification, reliability assessment, and remaining lifetime estimation. The algorithms used can include common machine learning algorithms such as Support Vector Machines (SVM), Decision Trees (DT), and Random Forests (RF); deep learning algorithms such as Deep Convolutional Neural Networks (CNN), Long Short-Term Memory Networks (LSTM), and Deep Autoencoders (DAE); or combinations of both, such as SVM+CNN, CNN+LSTM, etc., as well as various advanced fault diagnosis algorithms such as Physical Information Networks (PINN) and Transformers. These algorithms are not the subject of this application and will not be elaborated upon here. Users can choose appropriate algorithm combinations based on their actual diagnostic needs.

[0083] (2) Visualization processing module

[0084] The visualization module is primarily responsible for packaging the diagnostic process and final results into visual charts or files. Process data includes: the model used and its hyperparameters, the composition of the training and test sets, graphs showing changes in accuracy and loss during model training, and other intermediate results related to the method (such as feature maps of CNNs, distribution of intermediate variables in DAEs, etc.). Final result data includes: fault classification accuracy and confusion matrices under different labels, accuracy and comparison graphs of remaining lifetime estimation, and the model's uncertainty boundaries. The visualization module is responsible for collecting the above results and data and displaying them on the cloud server interface; it is also responsible for packaging all results into compressed files to return the diagnostic results to the edge computing system.

[0085] (3) Data / file transfer module

[0086] The data / file transfer module of the cloud-based intelligent fault early warning platform is primarily responsible for sending the packaged files compiled in step 3.2 back to the edge computing system via the SFTP protocol. This allows users to analyze the diagnostic results of the tested system at the edge, thereby improving the methods and parameter configurations for edge-side data preprocessing and feature extraction, and further optimizing the cloud-based diagnostic results. This edge-cloud interconnected data / file transfer method effectively improves the effectiveness of edge-cloud collaborative fault diagnosis and early warning.

[0087] (4) Integration and testing of cloud-based intelligent fault early warning platform

[0088] The fault diagnosis and early warning module, visualization processing module, and data / file transfer module were integrated and tested. All modules were encapsulated in a main function that executes periodically, achieving fully automated processing from fault diagnosis and early warning to result visualization and data / file transfer. Then, by sending fault data from the edge computing system to the cloud server, the functionality of the cloud-based intelligent fault early warning platform was tested.

[0089] In a verification embodiment, taking a bearing as an example, a 5-DoF dynamic model and a fault model of the bearing were constructed in Matlab / Simulink. A GPIO interface for signal acquisition of the Raspberry Pi edge computing system was developed, and a complete virtual test bench model was built. Simultaneously, in the edge computing system based on the Raspberry Pi hardware, modules such as mean filtering and time-frequency domain feature extraction were deployed. Finally, fault features and their corresponding tags were sent to the Alibaba Cloud platform via SFTP. Two diagnostic models, Support Vector Machine and Deep Convolutional Neural Network (CNN), were deployed on the cloud platform, thus completing the "end-edge-cloud" end-to-end intelligent diagnostic platform based on the virtual test bench. This enabled the testing and verification of edge-cloud collaborative communication and intelligent diagnostic algorithms in a virtual simulation environment. Figure 3Acceleration signals under different fault conditions generated by the virtual bearing test bench are given, where (a)-(d) represent the bearing acceleration response under normal, inner ring fault, outer ring fault and ball fault conditions, respectively. Figure 4 A confusion matrix diagram for fault classification is given, where the subscripts "B", "IR", and "OR" in the labels represent the location of the fault as the ball, inner ring, and outer ring, respectively; "7", "14", and "21" represent the size of the fault as 0.007 inches, 0.014 inches, and 0.021 inches, respectively; and "N_0" represents the normal state. Figure 5 The results of the remaining lifetime estimation are presented. It can be seen that the classification accuracy for different fault types is above 99%, and the root mean square error of the RUL estimation is also small. This demonstrates the effectiveness of the entire "edge-cloud" platform architecture.

[0090] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above 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.

[0091] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these modifications and improvements all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A smart fault diagnosis platform based on a virtual test bench across the entire end-to-end-to-cloud link, characterized in that, The platform includes: a virtual test bench system, an edge computing system, and a cloud-based intelligent fault early warning system; The virtual test bench system is used to perform dynamic modeling and fault type modeling on the system or component under test, and to simulate the response characteristics of the system or component under test under set boundary conditions and fault states, and to convert the response characteristics into electrical signals; wherein, the virtual test bench system includes: a dynamic model, a fault model and a hardware interface; The dynamic model is used to establish the kinematic and dynamic equations of the measured system or component under various degrees of freedom based on Newton's laws; the kinematic and dynamic equations are used to characterize the response characteristics of the system under different boundary conditions. The fault model is used to model wear and aging faults in the tested system or component. The hardware interface is used to convert the response characteristics of the tested system or component simulated by the dynamic model and the fault model under set conditions and fault conditions into electrical signals, and transmit the electrical signals to the edge computing system through the TCP / IP protocol; The edge computing system is used to preprocess and extract features from the acquired electrical signals, and transmit the extracted features to the cloud-based intelligent fault early warning system. It is also used to receive fault diagnosis, reliability analysis, or remaining lifetime estimation results transmitted from the cloud-based intelligent fault early warning system, allowing users to analyze the diagnostic results of the tested system or component at the edge, and improve the edge-side data preprocessing and feature extraction methods and parameter configurations based on the diagnostic results, further optimizing the cloud-based diagnostic results. The edge computing system includes a hardware module composed of an embedded system or edge computing components and a signal processing algorithm module running on the hardware module. The signal processing algorithm module includes a signal preprocessing submodule, a feature extraction submodule, and a feature data / file transfer submodule. The signal preprocessing submodule is used to preprocess the acquired electrical signals; The feature extraction submodule is used to extract fault features from the time-frequency domain of the preprocessed signal; The feature data / file transfer submodule is used to transmit the fault feature data extracted by the feature extraction submodule to the cloud-based intelligent fault early warning system. The cloud-based intelligent fault early warning system is used to perform fault diagnosis, reliability analysis, or remaining lifetime estimation on the tested system or component based on features extracted by the edge computing system, and transmit the results of the fault diagnosis, reliability analysis, or remaining lifetime estimation to the edge computing system; wherein, the cloud-based intelligent fault early warning system includes a fault diagnosis hardware module composed of a cloud computing platform or a high-performance server and a fault diagnosis algorithm module running on the fault diagnosis hardware module; the fault diagnosis algorithm module includes: a fault diagnosis and early warning submodule, a visualization processing submodule, and a fault data / file transfer submodule; The fault diagnosis and early warning submodule is used to reason and analyze the fault feature data sent by the edge computing system using machine learning methods, deep learning methods, or a combination of machine learning and deep learning methods, so as to realize fault classification, reliability assessment and remaining lifetime estimation of the tested system or component. The visualization processing submodule is used to package the process results and final results of the fault diagnosis and early warning submodule into a visual chart or file. The fault data / file transfer submodule is used to send the visualized charts or files back to the edge computing system via the SFTP protocol.

Images