Locomotive air source compressor fault detection method and device, storage medium and electronic equipment

Abstract

The invention provides a locomotive air source compressor fault detection method and device, a storage medium and electronic equipment, and relates to the technical field of artificial intelligence. The method comprises the following steps: acquiring bearing experiment table data; obtaining bearing independent fault data based on the bearing experiment table data; determining performance return and failure indexes according to the bearing experiment table data and the bearing independent fault data; determining a trained transfer learning model according to the bearing experiment table data, the bearing independent fault data and performance return and failure indexes; and performing fault detection on the acquired operation data of the locomotive air source compressor by using the trained transfer learning model, and determining a fault detection result. The working state of the compressor can be effectively evaluated, potential faults can be found in advance, the faults of the compressor can be rapidly detected and positioned, the fault occurrence probability is reduced, and the service life of equipment is prolonged.

Description

Background Art

[0002] The locomotive air source system is an important component of the railway locomotive. It is mainly used to provide the compressed air required by the train braking system and other equipment that requires compressed air. It is crucial to ensure the safe operation of the train. The core component of the locomotive air source system is the compressor, which is used to suck in the outside air and compress it to a certain pressure level, and then transport it to the subsequent processing equipment.

[0003] The working condition of the compressor is crucial to the locomotive air source system. If the compressor is not detected in the early stage of a fault, as the fault continues to worsen, it will pose a threat to the safe operation of the locomotive. In the prior art, fault detection of the locomotive air source system is usually discovered when the fault has already affected the locomotive air source system, and it is impossible to detect the compressor failure in the early stage.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present disclosure, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] The present disclosure provides a locomotive air source compressor fault detection method and device, and a storage medium electronic device, which at least to a certain extent overcome the problem that the related art cannot detect the compressor fault in its early stage.

[0006] Other features and advantages of the present disclosure will become apparent from the following detailed description, or may be learned in part by the practice of the present disclosure.

[0007] According to one aspect of the present disclosure, a method for detecting a fault of a locomotive air source compressor is provided, comprising:

[0008] Obtain bearing test bench data;

[0009] Based on the bearing test bench data, obtaining independent bearing fault data;

[0010] Determining performance regression and failure indicators based on the bearing test bench data and the bearing independent failure data;

[0011] Determining a trained transfer learning model based on the bearing test bench data, the bearing independent fault data, and the performance regression and failure indicators;

[0012] The trained transfer learning model is used to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result.

[0013] In some embodiments, obtaining bearing test bench data includes:

[0014] Using a fault test bench, measurement data of a bearing in a normal operating state and measurement data of a bearing in a fault state are obtained; the measurement data at least includes: a vibration signal, a noise signal, and a current and voltage signal; the fault types of the bearing at least include: an inner ring fault, an outer ring fault, a rolling element fault, a cage fault, and a mixed fault;

[0015] The bearing test bench data is determined based on the measurement data of the bearing in a normal operating state and the measurement data of the bearing in a fault state.

[0016] In some embodiments, based on the bearing test bench data, obtaining independent bearing fault data includes:

[0017] Based on the measurement data of the bearing in a fault state in the bearing test bench data, the corresponding bearing fault state is simulated on the rotor test bench to obtain independent fault data of the bearing of the rotor test bench in the corresponding bearing fault state; the rotor test bench runs with independent bearings; the independent bearing fault data includes: vibration signals, noise signals and current and voltage signals.

[0018] In some embodiments, determining performance regression and failure indicators based on the bearing test bench data and the bearing independent failure data includes:

[0019] Extracting time domain features, frequency domain features, time-frequency domain features, noise features, electrical features, and lubrication features based on the bearing test bench data and the bearing independent fault data;

[0020] Performance regression and failure indicators are established based on the time domain characteristics, frequency domain characteristics, time-frequency domain characteristics, noise characteristics, electrical characteristics, and lubrication characteristics.

[0021] In some embodiments, determining a trained transfer learning model based on the bearing test bench data, the bearing independent fault data, and the performance regression and failure indicators includes:

[0022] Using the bearing independent fault data as source domain data, pre-training a transfer learning model in the source domain data, and determining a pre-trained transfer learning model;

[0023] Using the measurement data of the bearing in the normal operating state in the bearing test bench data as the target domain data, fine-tuning the pre-trained transfer learning model in the target domain data, and determining the fine-tuned transfer learning model;

[0024] Using a fine-tuned transfer learning model for adversarial domain adaptation, the feature distributions in the source domain data and the target domain data are brought close to each other, and adversarial source domain data and adversarial target domain data are determined;

[0025] Assigning weights to source domain samples in the adversarial source domain data according to the performance regression and failure indicators to determine weighted source domain data;

[0026] Mapping the weighted source domain data and the adversarial target domain data to a common feature space to determine feature space mapping data;

[0027] The fine-tuned transfer learning model is trained according to the feature space mapping data to determine a trained transfer learning model.

[0028] In some embodiments, using the trained transfer learning model to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result includes:

[0029] Collecting the operating data of the locomotive air source compressor; the locomotive air source compressor operating data at least includes: vibration data, temperature data, noise data and current and voltage data;

[0030] deploying the trained transfer learning model to a locomotive;

[0031] The locomotive air source compressor operation data is input into the trained transfer learning model for fault detection to determine the fault detection result; the fault detection result includes: the bearing fault type and the bearing fault development.

[0032] According to another aspect of the present disclosure, a locomotive air source compressor fault detection method is also provided, comprising:

[0033] A bearing test bench data acquisition module is used to acquire bearing test bench data;

[0034] A bearing independent fault data acquisition module, used to acquire bearing independent fault data based on the bearing test bench data;

[0035] A performance regression and failure index determination module, used to determine the performance regression and failure index according to the bearing test bench data and the bearing independent fault data;

[0036] A model training module, for determining a trained transfer learning model based on the bearing test bench data, the bearing independent fault data, and the performance regression and failure indicators;

[0037] The fault detection module is used to use the trained transfer learning model to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result.

[0038] According to another aspect of the present disclosure, an electronic device is also provided, which includes: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned locomotive air source compressor fault detection methods by executing the executable instructions.

[0039] According to another aspect of the present disclosure, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the locomotive air source compressor fault detection method described in any one of the above is implemented.

[0040] According to another aspect of the present disclosure, a computer program product is also provided, including a computer program, wherein when the computer program is executed by a processor, the computer program implements any one of the above-mentioned locomotive air source compressor fault detection methods.

[0041] The locomotive air source compressor fault detection method and device, and storage medium electronic device provided in the embodiments of the present invention obtain bearing test bench data; based on the bearing test bench data, obtain bearing independent fault data; determine performance degradation and failure indicators based on the bearing test bench data and the bearing independent fault data; determine a trained transfer learning model based on the bearing test bench data, the bearing independent fault data, and the performance degradation and failure indicators; use the trained transfer learning model to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result. The disclosed embodiment obtains the bearing test bench data corresponding to the compressor on the test bench, and obtains the bearing independent fault data based on the bearing test bench data. Among the compressor faults, the bearing fault is the most significant and easy to detect. The performance regression and failure index are determined according to the bearing test bench data and the bearing independent fault data. The bearing fault development is evaluated by the performance regression and failure index. The transfer learning model is trained by the bearing test bench data, the bearing independent fault data, and the performance regression and failure index to obtain a trained transfer learning model. The transfer learning strategy is used to realize the migration from the experimental device to the locomotive air source system, and the locomotive air source compressor operation data is collected in real time, and input into the trained transfer learning model for fault detection to obtain the fault detection result. Under the complex operating conditions of the locomotive, the stability characteristics, correlation characteristics and variability characteristics in the fault characterization under complex conditions can be analyzed, and the working state of the compressor can be effectively evaluated, potential faults can be discovered in advance, and the rapid detection and positioning of the compressor fault can be realized, the probability of fault occurrence can be reduced, and the service life of the equipment can be extended. Based on performance regression and failure indicators, the fault type and severity can be judged more accurately, and unsupervised learning of compressor status detection and fault detection can be achieved through transfer learning, ensuring the reliable operation of the compressor under various working conditions and reporting faults to ensure the safe operation of railway locomotives.

[0042] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0044] Figure 1 A schematic diagram showing the system structure of a locomotive air source compressor fault detection method according to an embodiment of the present disclosure.

[0045] Figure 2 A schematic diagram of a locomotive air source compressor fault detection method in an embodiment of the present disclosure is shown.

[0046] Figure 3 A flow chart of a locomotive air source compressor fault detection method in an embodiment of the present disclosure is shown.

[0047] Figure 4 A schematic diagram of a locomotive air source compressor fault detection device in an embodiment of the present disclosure is shown.

[0048] Figure 5 A structural block diagram of a computer device for detecting a locomotive air source compressor fault in an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0049] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that the disclosure will be more comprehensive and complete and to fully convey the concepts of the example embodiments to those skilled in the art. The described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0050] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0051] The specific implementation of the embodiment of the present disclosure is described in detail below with reference to the accompanying drawings.

[0052] Figure 1 FIG. 2 shows an exemplary application system architecture diagram to which the locomotive air source compressor fault detection method according to the embodiment of the present disclosure can be applied. Figure 1 As shown, the system architecture may include a terminal device 101 , a network 102 and a server 103 .

[0053] The network 102 is a medium for providing a communication link between the terminal device 101 and the server 103, and can be a wired network or a wireless network.

[0054] Optionally, the wireless network or wired network described above uses standard communication technology and / or protocol. The network is usually the Internet, but it can also be any network, including but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or any combination of a virtual private network). In some embodiments, technologies and / or formats including Hyper Text Mark-up Language (HTML), Extensible Markup Language (XML), etc. are used to represent data exchanged through the network. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec) can also be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above data communication technologies.

[0055] The terminal device 101 can be various electronic devices, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, wearable devices, augmented reality devices, virtual reality devices, etc.

[0056] Optionally, the client of the application installed in different terminal devices 101 is the same, or the client of the same type of application based on different operating systems. Based on the different terminal platforms, the specific form of the client of the application can also be different, for example, the application client can be a mobile client, a PC client, etc.

[0057] The server 103 may be a server that provides various services, such as a background management server that provides support for the device operated by the user using the terminal device 101. The background management server may analyze and process the received request and other data, and feed back the processing results to the terminal device.

[0058] Optionally, the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. The terminal can be a smart phone, tablet computer, laptop computer, desktop computer, smart speaker, smart watch, etc., but is not limited to this. The terminal and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in this application.

[0059] Those skilled in the art will know that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration, and any number of terminal devices, networks and servers may be provided according to actual needs, and the embodiments of the present disclosure do not limit this.

[0060] Under the above system architecture, a locomotive air source compressor fault detection method is provided in an embodiment of the present disclosure, and the method can be executed by any electronic device with computing and processing capabilities.

[0061] In some embodiments, the locomotive air source compressor fault detection method provided in the embodiments of the present disclosure can be executed by the terminal device of the above-mentioned system architecture; in other embodiments, the locomotive air source compressor fault detection method provided in the embodiments of the present disclosure can be executed by the server in the above-mentioned system architecture; in other embodiments, the locomotive air source compressor fault detection method provided in the embodiments of the present disclosure can be implemented by the terminal device and the server in the above-mentioned system architecture through interaction.

[0062] Figure 2 A schematic diagram of a locomotive air source compressor fault detection method according to an embodiment of the present disclosure is shown. Figure 2 As shown, the locomotive air source compressor fault detection method provided in the embodiment of the present disclosure includes the following steps:

[0063] Step S202: Acquire bearing test bench data;

[0064] Step S204: acquiring independent bearing fault data based on the bearing test bench data;

[0065] Step S206: determining performance regression and failure indicators according to the bearing test bench data and the bearing independent failure data;

[0066] Step S208: determining a trained transfer learning model according to the bearing test bench data, the bearing independent fault data, and the performance regression and failure indicators;

[0067] Step S2010: Use the trained transfer learning model to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result.

[0068] The disclosed embodiment obtains the bearing test bench data corresponding to the compressor on the test bench, and obtains the bearing independent fault data based on the bearing test bench data. Among the compressor faults, the bearing fault is the most significant and easy to detect. The performance regression and failure index are determined according to the bearing test bench data and the bearing independent fault data. The bearing fault development is evaluated by the performance regression and failure index. The transfer learning model is trained by the bearing test bench data, the bearing independent fault data, and the performance regression and failure index to obtain a trained transfer learning model. The transfer learning strategy is used to realize the migration from the experimental device to the locomotive air source system, and the locomotive air source compressor operation data is collected in real time, and input into the trained transfer learning model for fault detection to obtain the fault detection result. Under the complex operating conditions of the locomotive, the stability characteristics, correlation characteristics and variability characteristics in the fault characterization under complex conditions can be analyzed, and the working state of the compressor can be effectively evaluated, potential faults can be discovered in advance, and the rapid detection and positioning of the compressor fault can be realized, the probability of fault occurrence can be reduced, and the service life of the equipment can be extended. Based on performance regression and failure indicators, the fault type and severity can be judged more accurately, and unsupervised learning of compressor status detection and fault detection can be achieved through transfer learning, ensuring the reliable operation of the compressor under various working conditions and reporting faults to ensure the safe operation of railway locomotives.

[0069] During the operation of the compressor, vibration signals are an important indicator of its health status. By monitoring and analyzing these vibration signals, the working status of the compressor can be effectively evaluated and potential faults can be discovered in advance.

[0070] In the embodiment, obtaining bearing test bench data includes:

[0071] Using the fault test bench, obtain the measurement data of the bearing in normal operation and the measurement data of the bearing in fault state; the measurement data at least include: vibration signal, noise signal and current and voltage signal; the fault types of the bearing at least include: inner ring fault, outer ring fault, rolling element fault, cage fault and mixed fault;

[0072] The bearing test bench data is determined based on the measurement data of the bearing in normal operation and the measurement data of the bearing in a fault state.

[0073] In the embodiment, the fault test bench is used to simulate the locomotive systems and components under various operating conditions to detect their performance, reliability and safety, and help identify potential failure modes. Different fault conditions, such as electrical faults, mechanical faults, etc., can be simulated to evaluate the response of the locomotive. By applying different loads and environmental conditions, the maximum working capacity and stability of locomotive components or systems are tested.

[0074] The compressor is placed on a fault test bench, and vibration signals, noise signals, and current and voltage signals of the compressor are obtained through a variety of sensors. Specifically, when the compressor is operating normally, the bearing is in a normal operating state, and measurement data of the bearing in a normal operating state is obtained, and the measurement data of the bearing in a normal operating state at least includes vibration signals, noise signals, and current and voltage signals.

[0075] Furthermore, the compressor is adjusted to different bearing fault types on the fault test bench, wherein the fault types include at least inner ring fault, outer ring fault, rolling element fault, cage fault and mixed fault. In each bearing fault type, the measurement data of the bearing in a faulty state is collected. The measurement data of the bearing in a faulty state also includes at least vibration signals, noise signals and current and voltage signals. The measurement data of the bearing in a normal operating state and the measurement data of the bearing in a faulty state are used to construct a basic general test bench bearing fault data set, which is used as a basic data set for various bearing fault data, providing data support for subsequent analysis and modeling. The above basic general test bench bearing fault data set is determined as the bearing test bench data, providing a comprehensive data basis for subsequent analysis.

[0076] In the embodiment, based on the bearing test bench data, obtaining the independent bearing fault data includes:

[0077] Based on the measurement data of the bearing in a fault state in the bearing test bench data, the corresponding bearing fault state is simulated on the rotor test bench to obtain independent fault data of the bearing in the corresponding bearing fault state of the rotor test bench; the rotor test bench runs with independent bearings; the independent bearing fault data includes: vibration signals, noise signals and current and voltage signals.

[0078] After obtaining the measurement data of the bearing in the fault state corresponding to different bearing fault types, the corresponding bearing fault state is simulated on the rotor test bench based on the bearing fault type. The rotor test bench is an important tool for simulating, testing and analyzing different fault states. For the simulation of bearing faults, this type of test bench can help engineers and technicians better understand the fault characteristics, thereby improving fault diagnosis capabilities and optimizing maintenance strategies. The rotor test bench is usually equipped with a high-performance drive system, precision sensors and data acquisition systems. These devices can accurately control parameters such as speed and load, and monitor the working status of the rotor and related components (such as bearings) in real time. In order to simulate the bearing faults that may occur in actual operation, some defects can be made on the independent bearings used on the rotor test bench. For example, the corresponding bearing fault type is simulated by drilling, cutting or wearing to obtain independent fault data of the bearing of the rotor test bench in the corresponding bearing fault state; further, different bearing fault types are adjusted to obtain independent fault data corresponding to different fault types respectively, and different bearing fault types are adjusted through the rotor test bench to obtain bearing fault data sets of independent components. The rotor test bench simulates various fault forms, such as bearing fault inner ring fault, outer ring fault, rolling element fault, cage fault, mixed fault, etc., and collects corresponding fault data. These independent bearing fault data also include vibration signals, temperature signals, noise signals, current and voltage signals, etc.

[0079] In the embodiment, the performance regression and failure indicators are determined based on the bearing test bench data and the bearing independent failure data, including:

[0080] According to the bearing test bench data and the bearing independent fault data, the time domain features, frequency domain features, time-frequency domain features, noise features, electrical features and lubrication features are extracted;

[0081] Performance regression and failure indicators are established based on time domain characteristics, frequency domain characteristics, time-frequency domain characteristics, noise characteristics, electrical characteristics and lubrication characteristics.

[0082] In actual engineering applications, analyzing the vibration signals of mechanical equipment is an important means to achieve equipment status monitoring and fault diagnosis. The vibration signals measured by compressors and dryers in engineering are generally time domain signals, which describe the changes of signals over time and can truly reflect the response characteristics of the equipment. As the fault occurs and develops, the signal frequency structure will also change. Therefore, time domain analysis and frequency domain analysis are the most important processing methods in vibration signal analysis. This embodiment collects and analyzes vibration signals to study the characteristics of fault characterization changes under the influence of multiple factors such as operating conditions and fault degree.

[0083] Specifically, according to the bearing test bench data and the bearing independent fault data, the time domain features are extracted, wherein the time domain features at least include: mean, variance, peak factor, kurtosis, etc. According to the bearing test bench data and the bearing independent fault data, the frequency domain features are extracted, wherein the frequency domain features at least include: spectrum peak, frequency band energy, spectrum centroid, etc. According to the bearing test bench data and the bearing independent fault data, the time-frequency domain features are extracted, wherein the time-frequency domain features at least include: such as short-time Fourier transform, wavelet transform, etc. According to the bearing test bench data and the bearing independent fault data, the noise features are extracted, wherein the noise features at least include: sound pressure level, noise spectrum, etc. According to the bearing test bench data and the bearing independent fault data, the electrical features are extracted, wherein the electrical features at least include: current signal, voltage signal, etc. According to the bearing test bench data and the bearing independent fault data, the lubrication features are extracted, wherein the lubrication features at least include: lubricating oil quality, temperature, etc.

[0084] Based on time domain characteristics, frequency domain characteristics, time-frequency domain characteristics, noise characteristics, electrical characteristics and lubrication characteristics, performance regression and failure indicators are established. A comprehensive evaluation system is built to monitor and evaluate the operating status and health of bearings. The effectiveness of each indicator is verified through experiments to ensure that it can accurately reflect the state changes of the compressor.

[0085] In the embodiment, the trained transfer learning model is determined based on the bearing test bench data, the bearing independent fault data, and the performance regression and failure indicators, including:

[0086] Using the bearing independent fault data as source domain data, pre-training the transfer learning model in the source domain data, and determining the pre-trained transfer learning model;

[0087] The measurement data of the bearing in the normal operating state in the bearing test bench data is used as the target domain data, and the pre-trained transfer learning model is fine-tuned in the target domain data to determine the fine-tuned transfer learning model;

[0088] Using a fine-tuned transfer learning model for adversarial domain adaptation, the feature distributions in the source domain data and the target domain data are brought close to each other, and adversarial source domain data and adversarial target domain data are determined;

[0089] Assigning weights to source domain samples in the adversarial source domain data according to performance regression and failure indicators to determine weighted source domain data;

[0090] Mapping the weighted source domain data and the adversarial target domain data to a common feature space, and determining feature space mapping data;

[0091] The fine-tuned transfer learning model is trained according to the feature space mapping data to determine the trained transfer learning model.

[0092] In the embodiment, a migration strategy is formulated, and effective online migration from the experimental device to the locomotive air source system is achieved through transfer learning technology.

[0093] The bearing independent fault data is used as the source domain data, and the transfer learning model is pre-trained in the source domain data to determine the pre-trained transfer learning model; the transfer learning model is pre-trained using the source domain data, with the goal of allowing the model to learn some common feature representations that are useful in many tasks.

[0094] The measurement data of the bearing in normal operating state in the bearing test bench data is used as the target domain data, and the pre-trained transfer learning model is fine-tuned in the target domain data to determine the fine-tuned transfer learning model.

[0095] Fine-tune the pre-trained transfer learning model using the target domain data. This step usually uses a lower learning rate to avoid drastically modifying the learned knowledge. The fine-tuning process can adjust the model structure according to the characteristics of the target task, such as adding or deleting certain layers to better suit specific task requirements.

[0096] Using a fine-tuned transfer learning model for adversarial domain adaptation, the feature distributions in the source domain data and the target domain data are made close, and the adversarial source domain data and the adversarial target domain data are determined; specifically, a domain classifier is constructed, the purpose of which is to distinguish whether the sample is from the source domain or the target domain. At the same time, during the training process, the feature extractor is optimized so that it can generate feature representations that confuse the domain classifier, so that the feature distributions of the source domain and the target domain are as close as possible; adversarial training is achieved by alternately training the domain classifier and the feature extractor. On the one hand, the ability of the domain classifier to distinguish between the source domain and the target domain is improved as much as possible; on the other hand, the feature extractor is optimized to minimize this ability to distinguish, so as to achieve the purpose of domain-invariant feature learning. Through adversarial domain adaptation technology, the feature distributions of the source domain and the target domain are made as close as possible.

[0097] Assign weights to source domain samples in the adversarial source domain data according to performance regression and failure indicators, so that the distribution of weighted source domain data is closer to the distribution of target domain data, thereby obtaining weighted source domain data;

[0098] Assign weights to source domain samples based on performance regression and failure indicators, such as calculating the distance or similarity between each source domain sample and the target domain sample. The higher the weight, the more important the sample is, and the greater its role in narrowing the gap between the source domain and the target domain. Introduce a weighted loss function during model training to adjust the contribution of each sample to the overall loss based on the assigned weight. This can make the source domain data distribution closer to the target domain data distribution to a certain extent, thereby improving the performance of the model on the target domain.

[0099] Through feature mapping technology, the weighted source domain data and the adversarial target domain data are mapped to a common feature space to determine the feature space mapping data; specifically, a transformation or mapping method is learned to map the source domain and target domain data to a common feature space. In the common feature space, certain constraints are imposed to ensure that the data from the two different domains have similar distributions in this space. For example, this can be achieved by minimizing the distribution differences between the two domains in the common feature space, so that the model can effectively transfer the knowledge learned from the source domain to the target domain.

[0100] In an embodiment, the fine-tuned transfer learning model is trained according to the feature space mapping data, and when the transfer learning model reaches a convergence condition, the training is stopped and the trained transfer learning model is output.

[0101] In the embodiment, the trained transfer learning model is used to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result, including:

[0102] Collecting the operating data of the locomotive air source compressor; the operating data of the locomotive air source compressor includes at least: vibration data, temperature data, noise data and current and voltage data;

[0103] Deploy the trained transfer learning model to the locomotive;

[0104] The locomotive air source compressor operation data is input into the trained transfer learning model for fault detection to determine the fault detection results; the fault detection results include: the bearing fault type and the bearing fault development.

[0105] Install a variety of sensors on railway locomotives to collect real-time operating data of the locomotive air source compressor, including at least vibration data, temperature data, noise data, and current and voltage data; deploy the trained transfer learning model to the locomotive to achieve online migration diagnosis from the experimental device to the equipment. Input the locomotive air source compressor operating data into the trained transfer learning model for fault detection to determine the fault detection results; the fault detection results include: bearing fault type and bearing fault development. Using the locomotive air source compressor operating data obtained by online monitoring, the compressor bearing fault is detected and diagnosed in real time through the transfer learning model. According to the fault diagnosis results, timely issue early warning signals to guide maintenance personnel to perform inspections and maintenance, thereby achieving efficient and reliable online fault diagnosis.

[0106] By collecting and analyzing vibration signals, we study the changing characteristics of fault characterization under the influence of multiple factors such as operating conditions and fault degree, and obtain the stability characteristics, correlation characteristics and variability characteristics of fault characterization under complex working conditions, thereby obtaining a fault identification scheme for key components of the locomotive air source system based on vibration signal analysis.

[0107] It can realize the rapid detection and location of compressor faults, reduce the probability of faults, and extend the service life of the equipment. Multivariate statistical analysis methods, such as principal component analysis and independent component analysis, can further improve the accuracy of fault detection and help engineers more accurately determine the type and severity of faults. The comprehensive application of vibration signal acquisition, unsupervised learning and multivariate statistical analysis methods can provide strong technical support for compressor status monitoring and fault detection, ensuring the reliable operation of compressors under various working conditions.

[0108] Figure 3 A flow chart of a method for detecting a fault of a locomotive air source compressor according to an embodiment of the present disclosure is shown. Figure 3 As shown in the figure, the main process steps include: establishing a basic general test bench bearing fault data set, adjusting different bearing fault types through the rotor test bench, obtaining the bearing fault data set of independent components, and establishing the corresponding performance degradation and failure index system. Then, through the transfer learning technology, taking the normal state of the compressor as the target domain and the independent bearing fault as the source domain, a migration strategy is formulated to achieve online migration diagnosis from the experimental device to the equipment. Using the fault test bench, normal and different fault types (such as inner ring fault, outer ring fault, rolling element fault, cage fault, and mixed fault) of the bearing operation data are collected. Through the rotor test bench, different bearing fault types are adjusted to obtain the bearing fault data set of independent components. A performance degradation and failure index system is established, and a comprehensive evaluation system is constructed to monitor and evaluate the operating status and health of the bearing. A migration strategy is formulated to achieve effective migration from the experimental device to the equipment online through the transfer learning technology. A migration diagnosis from the experimental device to the equipment online is achieved. Sensors are installed on the actual equipment (such as the compressor) to collect vibration, temperature, noise, current and voltage signals in real time.

[0109] It should be noted that the acquisition, storage, use, and processing of data in the technical solution disclosed in this disclosure are in compliance with the relevant provisions of national laws and regulations, and various types of data such as personal identity data, operation data, behavioral data, etc. related to individuals, customers, and groups obtained in the embodiments of this disclosure have all been authorized.

[0110] Based on the same inventive concept, the present disclosure also provides a locomotive air source compressor fault detection device, as described in the following embodiments. Since the principle of solving the problem in the device embodiment is similar to that in the above method embodiment, the implementation of the device embodiment can refer to the implementation of the above method embodiment, and the repeated parts will not be repeated.

[0111] Figure 4 FIG. 1 is a schematic diagram of a locomotive air source compressor fault detection device according to an embodiment of the present disclosure. Figure 4 As shown, the device comprises:

[0112] A bearing test bench data acquisition module 401 is used to acquire bearing test bench data;

[0113] A bearing independent fault data acquisition module 402 is used to acquire bearing independent fault data based on the bearing test bench data;

[0114] The performance degradation and failure index determination module 403 is used to determine the performance degradation and failure index according to the bearing test bench data and the bearing independent fault data;

[0115] A model training module 404 is used to determine a trained transfer learning model based on the bearing test bench data, the bearing independent fault data, and the performance regression and failure indicators;

[0116] The fault detection module 405 is used to use the trained transfer learning model to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result.

[0117] It should be noted that the bearing test bench data acquisition module 401, the bearing independent fault data acquisition module 402, the performance regression and failure index determination module 403, the model training module 404 and the fault detection module 405 correspond to S202 to S2010 in the method embodiment, and the examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above method embodiment. It should be noted that the above modules as part of the device can be executed in a computer system such as a set of computer executable instructions.

[0118] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0119] Refer to the following Figure 5The electronic device 500 according to this embodiment of the present disclosure is described. Figure 5 The electronic device 500 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0120] like Figure 5 As shown, the electronic device 500 is in the form of a general computing device. The components of the electronic device 500 may include but are not limited to: at least one processing unit 510, at least one storage unit 520, and a bus 530 connecting different system components (including the storage unit 520 and the processing unit 510).

[0121] The storage unit stores a program code, and the program code can be executed by the processing unit 510, so that the processing unit 510 executes the steps described in the above "Exemplary Method" section of this specification according to various exemplary embodiments of the present disclosure. For example, the processing unit 510 can execute the following steps of the above method embodiment: obtaining bearing test bench data; obtaining bearing independent fault data based on the bearing test bench data; determining performance regression and failure indicators based on the bearing test bench data and the bearing independent fault data; determining a trained transfer learning model based on the bearing test bench data, the bearing independent fault data, and the performance regression and failure indicators; using the trained transfer learning model to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result.

[0122] The storage unit 520 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 5201 and / or a cache storage unit 5202 , and may further include a read-only storage unit (ROM) 5203 .

[0123] The storage unit 520 may also include a program / utility 5204 having a set (at least one) of program modules 5205, such program modules 5205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0124] Bus 530 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0125] The electronic device 500 may also communicate with one or more external devices 540 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 500, and / or communicate with any device that enables the electronic device 500 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 550. Furthermore, the electronic device 500 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 560. As shown, the network adapter 560 communicates with other modules of the electronic device 500 via a bus 530. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 500, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0126] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0127] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart may be implemented as a computer program product, which includes: a computer program, which implements the above locomotive air source compressor fault detection method when executed by a processor.

[0128] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, which may be a readable signal medium or a readable storage medium. A program product capable of implementing the above method of the present disclosure is stored thereon. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to cause the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above “Exemplary Method” section of this specification.

[0129] More specific examples of computer-readable storage media in the present disclosure may include, but are not limited to, an electrical connection having one or more conductors, 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), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0130] In the present disclosure, a computer readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, wherein a readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A readable signal medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0131] Alternatively, the program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0132] In a specific implementation, the program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0133] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0134] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0135] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the method according to the implementation of the present disclosure.

[0136] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present disclosure is intended to cover any variations, uses or adaptations of the present disclosure, which follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the appended claims.

Claims

1. A locomotive air source compressor fault detection method, characterized in that: include: Obtain bearing test bench data; Based on the bearing test bench data, obtaining independent bearing fault data; Determining performance regression and failure indicators based on the bearing test bench data and the bearing independent failure data; Determining a trained transfer learning model based on the bearing test bench data, the bearing independent fault data, and the performance regression and failure indicators; The trained transfer learning model is used to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result.

2. The locomotive air source compressor fault detection method according to claim 1, characterized in that: Obtain bearing test bench data, including: Using a fault test bench, measurement data of a bearing in a normal operating state and measurement data of a bearing in a fault state are obtained; the measurement data at least includes: a vibration signal, a noise signal, and a current and voltage signal; the fault types of the bearing at least include: an inner ring fault, an outer ring fault, a rolling element fault, a cage fault, and a mixed fault; The bearing test bench data is determined based on the measurement data of the bearing in a normal operating state and the measurement data of the bearing in a fault state.

3. The locomotive air source compressor fault detection method according to claim 2, characterized in that: Based on the bearing test bench data, independent bearing fault data is obtained, including: Based on the measurement data of the bearing in a fault state in the bearing test bench data, the corresponding bearing fault state is simulated on the rotor test bench to obtain independent fault data of the bearing of the rotor test bench in the corresponding bearing fault state; the rotor test bench runs with independent bearings; the independent bearing fault data includes: vibration signals, noise signals and current and voltage signals.

4. The locomotive air source compressor fault detection method according to claim 1, characterized in that: According to the bearing test bench data and the bearing independent fault data, performance regression and failure indicators are determined, including: Extracting time domain features, frequency domain features, time-frequency domain features, noise features, electrical features, and lubrication features based on the bearing test bench data and the bearing independent fault data; Performance regression and failure indicators are established based on the time domain characteristics, frequency domain characteristics, time-frequency domain characteristics, noise characteristics, electrical characteristics, and lubrication characteristics.

5. The locomotive air source compressor fault detection method according to claim 3, characterized in that: Based on the bearing test bench data, the bearing independent fault data, and the performance regression and failure indicators, a trained transfer learning model is determined, including: Using the bearing independent fault data as source domain data, pre-training a transfer learning model in the source domain data, and determining a pre-trained transfer learning model; Using the measurement data of the bearing in the normal operating state in the bearing test bench data as the target domain data, fine-tuning the pre-trained transfer learning model in the target domain data, and determining the fine-tuned transfer learning model; Using a fine-tuned transfer learning model for adversarial domain adaptation, the feature distributions in the source domain data and the target domain data are brought close to each other, and adversarial source domain data and adversarial target domain data are determined; Assigning weights to source domain samples in the adversarial source domain data according to the performance regression and failure indicators to determine weighted source domain data; Mapping the weighted source domain data and the adversarial target domain data to a common feature space to determine feature space mapping data; The fine-tuned transfer learning model is trained according to the feature space mapping data to determine a trained transfer learning model.

6. The locomotive air source compressor fault detection method according to claim 1, characterized in that: The trained transfer learning model is used to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result, including: Collecting the operating data of the locomotive air source compressor; the locomotive air source compressor operating data at least includes: vibration data, temperature data, noise data and current and voltage data; deploying the trained transfer learning model to a locomotive; The locomotive air source compressor operation data is input into the trained transfer learning model for fault detection to determine the fault detection result; the fault detection result includes: the bearing fault type and the bearing fault development.

7. A locomotive air source compressor fault detection method, characterized in that: include: A bearing test bench data acquisition module is used to acquire bearing test bench data; A bearing independent fault data acquisition module, used to acquire bearing independent fault data based on the bearing test bench data; A performance regression and failure index determination module, used to determine the performance regression and failure index according to the bearing test bench data and the bearing independent fault data; A model training module, for determining a trained transfer learning model based on the bearing test bench data, the bearing independent fault data, and the performance regression and failure indicators; The fault detection module is used to use the trained transfer learning model to perform fault detection on the collected locomotive air source compressor operation data to determine the fault detection result.

8. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; Wherein, the processor is configured to execute the locomotive air source compressor fault detection method according to any one of claims 1 to 6 by executing the executable instructions.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the locomotive air source compressor fault detection method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the locomotive air source compressor fault detection method according to any one of claims 1 to 6 is implemented.

Images