Vibration fault detection model training method, vibration fault detection method and device

Abstract

The invention provides a vibration fault detection model training method and device and a vibration fault detection method and device.The method comprises the steps that a first vibration signal of a gas turbine rotor bearing in the first direction, a second vibration signal of the gas turbine rotor bearing in the second direction and a sample noise signal are obtained; the first direction and the second direction are orthogonal; based on the first vibration signal, the second vibration signal and the sample noise signal, generating a sample RGB image, and obtaining a fault type label corresponding to the sample RGB image; based on the sample RGB image and the corresponding fault type label, training the to-be-trained model to obtain a vibration fault detection model; wherein the vibration fault detection model is used for detecting the vibration fault type of the gas turbine. The vibration signals and the noise signals in the two orthogonal directions are converted into the RGB images, the image expressive force can be enhanced, the model can capture richer and more detailed features, and the adaptive capacity of the model to noise and the fault type detection precision are improved.

Description

Technical Field

[0001] The present application relates to the technical field of fault detection, and in particular to a vibration fault detection model training method, a vibration fault detection method and a device. Background Art

[0002] Gas turbines are key equipment and are widely used in aviation, power generation and industrial fields. However, their complex structure and harsh working environment can easily cause vibration failures.

[0003] In related technologies, the detection method of vibration faults mainly relies on the extraction of time domain, frequency domain or time-frequency domain features, which has problems such as insufficient information utilization and insufficient diagnostic accuracy. Summary of the Invention

[0004] The present application aims to solve one of the technical problems in the related art at least to a certain extent.

[0005] To this end, the first purpose of this application is to propose a vibration fault detection model training method.

[0006] The second objective of this application is to provide a vibration fault detection method.

[0007] The third objective of this application is to provide a vibration fault detection model training device.

[0008] The fourth objective of this application is to provide a vibration fault detection device.

[0009] The fifth objective of this application is to provide an electronic device.

[0010] A sixth object of this application is to provide a computer-readable storage medium.

[0011] The seventh object of this application is to provide a computer program product.

[0012] To achieve the above objectives, the first embodiment of the present application proposes a vibration fault detection model training method, comprising:

[0013] Acquire a first vibration signal of a gas turbine rotor bearing in a first direction, a second vibration signal in a second direction, and a sample noise signal; the first direction and the second direction are orthogonal;

[0014] generating a sample RGB image based on the first vibration signal, the second vibration signal, and the sample noise signal, and obtaining a fault type label corresponding to the sample RGB image;

[0015] Based on the sample RGB images and the corresponding fault type labels, the to-be-trained model is trained to obtain a vibration fault detection model; wherein the vibration fault detection model is used to detect the vibration fault type of the gas turbine.

[0016] To achieve the above objectives, a second embodiment of the present application provides a vibration fault detection method, comprising:

[0017] Acquire a first real-time vibration signal of a gas turbine rotor bearing in a first direction, a second real-time vibration signal in a second direction, and a real-time noise signal; wherein the first direction and the second direction are orthogonal;

[0018] generating a target RGB image based on the first real-time vibration signal, the second real-time vibration signal, and the real-time noise signal;

[0019] The target RGB image is input into a trained vibration fault detection model to obtain the vibration fault type output by the vibration fault detection model; wherein the vibration fault detection model is trained based on sample RGB images and corresponding fault type labels.

[0020] To achieve the above objectives, the third embodiment of the present application proposes a vibration fault detection model training device, comprising:

[0021] an acquisition module, configured to acquire a first vibration signal of a gas turbine rotor bearing in a first direction, a second vibration signal in a second direction, and a sample noise signal; the first direction and the second direction being orthogonal;

[0022] a generating module, configured to generate a sample RGB image based on the first vibration signal, the second vibration signal, and the sample noise signal, and to obtain a fault type label corresponding to the sample RGB image;

[0023] A training module is used to train a to-be-trained model based on the sample RGB image and the corresponding fault type label to obtain a vibration fault detection model; wherein the vibration fault detection model is used to detect the vibration fault type of the gas turbine.

[0024] To achieve the above objectives, a fourth embodiment of the present application provides a vibration fault detection device, comprising:

[0025] an acquisition module, configured to acquire a first real-time vibration signal of a gas turbine rotor bearing in a first direction, a second real-time vibration signal in a second direction, and a real-time noise signal; wherein the first direction and the second direction are orthogonal;

[0026] a generating module, configured to generate a target RGB image based on the first real-time vibration signal, the second real-time vibration signal, and the real-time noise signal;

[0027] The detection module is used to input the target RGB image into a trained vibration fault detection model to obtain the vibration fault type output by the vibration fault detection model; wherein the vibration fault detection model is trained based on sample RGB images and corresponding fault type labels.

[0028] To achieve the above-mentioned purpose, the fifth aspect of the present application proposes an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the vibration fault detection model training method described in the first aspect of the embodiment of the present application or the vibration fault detection method described in the second aspect.

[0029] To achieve the above-mentioned purpose, the sixth embodiment of the present application proposes a computer-readable storage medium, which stores computer execution instructions. When the computer execution instructions are executed by the processor, they are used to implement the vibration fault detection model training method described in the first aspect of the embodiment of the present application or the vibration fault detection method described in the second aspect.

[0030] To achieve the above-mentioned purpose, the seventh embodiment of the present application proposes a computer program product, including a computer program, which, when executed by a processor, implements the vibration fault detection model training method described in the first aspect of the embodiment of the present application or the vibration fault detection method described in the second aspect.

[0031] The technical solution provided by this application brings at least the following beneficial effects:

[0032] By obtaining a first vibration signal of a gas turbine rotor bearing in a first direction, a second vibration signal in a second direction, and a sample noise signal; the first direction and the second direction are orthogonal; based on the first vibration signal, the second vibration signal and the sample noise signal, a sample RGB image is generated, and a fault type label corresponding to the sample RGB image is obtained; based on the sample RGB image and the corresponding fault type label, a training model is trained to obtain a vibration fault detection model; wherein the vibration fault detection model is used to detect the vibration fault type of the gas turbine. The present application converts the vibration signals and noise signals in two orthogonal directions into RGB images, which can effectively integrate the information of the vibration signal and the noise interference, enhance the image expression, and enable the model to capture richer and more detailed features; training the model based on the sample RGB image can enable the model to automatically identify and classify different faults, thereby processing more types of fault modes and making accurate predictions, with higher robustness and automation capabilities; in addition, using the noise signal as a channel of the RGB image improves the model's adaptability to noise, and the model can automatically eliminate interference and enhance the robustness of fault detection.

[0033] Additional aspects and advantages of the present application will be given in part in the description below, and in part will become apparent from the description below, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0035] Figure 1 A flowchart of a vibration fault detection model training method provided in one embodiment of the present application;

[0036] Figure 2 A flowchart of a vibration fault detection model training method provided by another embodiment of the present application;

[0037] Figure 3 A schematic flow chart of a vibration fault detection method provided in one embodiment of the present application;

[0038] Figure 4 A schematic structural diagram of a vibration fault detection model training device provided in one embodiment of the present application;

[0039] Figure 5 A schematic structural diagram of a vibration fault detection device provided in one embodiment of the present application;

[0040] Figure 6 This is a block diagram of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0041] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0042] The following describes the vibration fault detection model training method, vibration fault detection method and device according to the embodiments of the present application with reference to the accompanying drawings.

[0043] Figure 1 A flowchart of a vibration fault detection model training method provided in an embodiment of the present application.

[0044] like Figure 1 As shown, the vibration fault detection model training method includes the following steps:

[0045] Step 101: Acquire a first vibration signal of a gas turbine rotor bearing in a first direction, a second vibration signal in a second direction, and a sample noise signal; the first direction and the second direction are orthogonal;

[0046] Step 102: generating a sample RGB image based on the first vibration signal, the second vibration signal, and the sample noise signal, and obtaining a fault type label corresponding to the sample RGB image;

[0047] Step 103 : Based on the sample RGB images and the corresponding fault type labels, the to-be-trained model is trained to obtain a vibration fault detection model; wherein the vibration fault detection model is used to detect the vibration fault type of the gas turbine.

[0048] In some embodiments of the present application, an acceleration sensor may be used to collect gas turbine rotor bearing vibration data, including vibration signals in two orthogonal directions.

[0049] In some embodiments of the present application, the sample noise signal may refer to an environmental noise signal or a random white noise signal.

[0050] In some embodiments of the present application, obtaining a sample noise signal includes: filtering and decomposing the first vibration signal and the second vibration signal respectively to obtain a first noise signal in the first vibration signal and a second noise signal in the second vibration signal; and obtaining the sample noise signal based on the first noise signal and the second noise signal.

[0051] For example, a signal decomposition algorithm, such as a wavelet transform algorithm, can be used to decompose the first vibration signal into a first trend signal and a first noise signal, and to decompose the second vibration signal into a second trend signal and a second noise signal. The first trend signal and the second trend signal can be understood as signals that have been filtered.

[0052] The first noise signal and the second noise signal correspond to multiple moments, and each moment corresponds to a signal value.

[0053] Therefore, obtaining a sample noise signal based on the first noise signal and the second noise signal can accurately combine the actual vibration noise conditions of the gas turbine rotor bearing in the orthogonal direction, making the sample noise signal more targeted, and helping to generate a sample RGB image that is more in line with the actual scene, thereby improving the training effect of the vibration fault detection model and the accuracy of fault type detection.

[0054] In some embodiments of the present application, as a possible implementation manner, any one of the first noise signal and the second noise signal may be used as a sample noise signal.

[0055] In some embodiments of the present application, as another possible implementation method, for any target moment among multiple moments, a first signal value of the target moment in the first noise signal and a second signal value of the target moment in the second noise signal are obtained; the first signal value and the second signal value are averaged to obtain a mean signal value; and a sample noise signal is generated based on the mean signal value corresponding to each moment among the multiple moments.

[0056] The signal value in the first noise signal and the signal value in the second noise signal at each moment may be averaged to obtain the average signal value corresponding to each moment, so as to obtain the sample noise signal.

[0057] Therefore, by fusing the noise signal values ​​at the target moment in the orthogonal direction and balancing the influence of noise in different directions through mean processing, the generated sample noise signal can be made more representative and stable, providing more reliable training data for model training, and further improving the model training effect and the accuracy of vibration fault detection.

[0058] In some embodiments of the present application, as another possible implementation, for any target moment among multiple moments, a target signal value can be obtained from the first signal value at the target moment in the first noise signal and the second signal value at the target moment in the second noise signal; based on the target signal value corresponding to each moment among the multiple moments, a sample noise signal is generated. Wherein, the target signal value is the maximum signal value or the minimum signal value

[0059] In some embodiments of the present application, a sample RGB image is generated based on the first vibration signal, the second vibration signal and the sample noise signal, including: splitting the first vibration signal, the second vibration signal and the sample noise signal according to the set image length to obtain corresponding multiple first vibration sub-signals, multiple second vibration sub-signals and multiple target noise sub-signals; according to the time sequence, taking the signal value corresponding to each first vibration sub-signal as a row of pixel values ​​in the first channel to obtain a first channel image; taking the signal value corresponding to each second vibration sub-signal as a row of pixel values ​​in the second channel to obtain a second channel image; taking the signal value corresponding to each target noise sub-signal as a row of pixel values ​​in the third channel to obtain a third channel image; generating a sample RGB image based on the first channel image, the second channel image and the third channel image.

[0060] Among them, the signal lengths of the first vibrator signal, the second vibrator signal and the target noise sub-signal are the same as the set image length, that is, when the set image length is N, the first vibrator signal, the second vibrator signal and the target noise sub-signal respectively include N signal values.

[0061] For example, the set image length can be 28, that is, the size of the sample RGB image is 28*28; the set image length can also be 224, that is, the size of the sample RGB image is 224*224. It should be noted that the number of sample RGB images is multiple.

[0062] The first channel, the second channel, and the third channel may correspond to red (R), green (G), and blue (B) channels respectively in any order to form an RGB channel combination.

[0063] Therefore, splitting the vibration signal and noise signal in orthogonal directions into pixel rows can completely preserve the signal timing characteristics and noise information. The generated sample RGB image is integrated with multi-dimensional data, which helps to improve the vibration fault detection model's recognition ability and accuracy for complex fault modes.

[0064] In some embodiments of the present application, the first vibration signal, the second vibration signal and the sample noise signal are split according to the set image length, respectively, to obtain corresponding multiple first vibration sub-signals, multiple second vibration sub-signals and multiple target noise sub-signals, including: splitting the first vibration signal according to the set image length to obtain multiple first sub-signals, splitting the second vibration signal to obtain multiple second sub-signals, and splitting the sample noise signal to obtain multiple third sub-signals; for any target sub-signal among the first sub-signal, the second sub-signal and the third sub-signal, normalizing the target sub-signal to obtain a corresponding normalized signal; mapping the normalized signal based on the set mapping method to obtain a mapping signal; wherein the signal value in the mapping signal is within the set value range; taking the mapping signal corresponding to the first sub-signal as the first vibration sub-signal, the mapping signal corresponding to the second sub-signal as the second vibration sub-signal, and the mapping signal corresponding to the third sub-signal as the target noise sub-signal.

[0065] When performing signal splitting, the signal can be split along the time axis according to the set image length; the mapping method is set to set the value range of the normalized signal value mapping value. For example, the set value range can be [0, 255].

[0066] Therefore, the signal is split according to the set image length and normalized and mapped, which unifies the data range of different signals. This not only facilitates subsequent image generation and processing, but also enables the pixel values ​​of each channel to be within a reasonable value range, enhances the consistency of image information, helps the model capture key features, and improves the accuracy and reliability of gas turbine vibration fault detection.

[0067] In some embodiments of the present application, obtaining a fault type label corresponding to a sample RGB image includes: performing time-frequency analysis on the sample RGB image using a fast Fourier transform algorithm or a wavelet transform algorithm; and determining a fault type label corresponding to the sample RGB image based on the time-frequency analysis result.

[0068] The fault type label corresponding to the sample RGB image may refer to a single fault type label or multiple fault type labels.

[0069] Therefore, using fast Fourier transform or wavelet transform algorithm for time-frequency analysis can deeply explore the frequency characteristics of vibration signals in sample RGB images, accurately locate faults, and make the determined fault type labels more accurate, thereby effectively improving the training quality of vibration fault detection models and fault diagnosis accuracy.

[0070] In some embodiments of the present application, the model to be trained may refer to a deep learning model, such as a convolutional neural network model. It should be noted that after the model to be trained is trained using a large-scale dataset, the model performance can be evaluated through cross-validation and test sets.

[0071] In an embodiment of the present application, a first vibration signal in a first direction, a second vibration signal in a second direction, and a sample noise signal of a gas turbine rotor bearing are obtained; the first direction and the second direction are orthogonal; a sample RGB image is generated based on the first vibration signal, the second vibration signal, and the sample noise signal, and a fault type label corresponding to the sample RGB image is obtained; based on the sample RGB image and the corresponding fault type label, a training model is trained to obtain a vibration fault detection model; wherein the vibration fault detection model is used to detect the vibration fault type of the gas turbine. The present application converts the vibration signals and the noise signal in two orthogonal directions into RGB images, which can effectively integrate the information of the vibration signal and the noise interference, enhance the image expression, and enable the model to capture richer and more detailed features; training the model based on the sample RGB image can enable the model to automatically identify and classify different faults, thereby processing a wider range of fault modes and making accurate predictions, with higher robustness and automation capabilities; in addition, using the noise signal as a channel of the RGB image improves the model's adaptability to noise, and the model can automatically eliminate interference, enhancing the robustness of fault detection.

[0072] This embodiment provides a vibration fault detection model training method. Figure 2 A flowchart of a vibration fault detection model training method provided in an embodiment of the present application.

[0073] like Figure 2 As shown, the vibration fault detection model training method may include the following steps:

[0074] Data measurement: For gas turbine rotor bearings, collect X-axis vibration signals and Y-axis vibration signals (i.e., first vibration signals and second vibration signals), and obtain target noise signals;

[0075] Image generation: crop (split), normalize, and map the collected signal to generate a sample RGB image;

[0076] Model training: Annotate sample RGB images to obtain corresponding fault type labels; train the model based on sample RGB images and corresponding fault type labels.

[0077] Among them, the trained vibration fault detection model can be used to detect the vibration fault type of gas turbines.

[0078] Compared to the related art method of using one-dimensional signals or two-dimensional feature maps for fault detection, this application constructs an end-to-end RGB image based on two orthogonal vibration signals and noise signals. This can simultaneously utilize the original information of multiple dimensions, retain implicit features, avoid information loss during data conversion, and increase the expressiveness of the data. In addition, by fusing the vibration and noise signals in two orthogonal directions, not only the robustness of the image is improved, but also the differentiation of different types of faults is enhanced. The introduction of noise signals enables the model to adapt to various interferences in the real environment, improving the accuracy of fault detection in practical applications.

[0079] This embodiment provides a vibration fault detection method. Figure 3 A flow chart of a vibration fault detection method provided in an embodiment of the present application.

[0080] like Figure 3 As shown, the vibration fault detection method may include the following steps:

[0081] Step 301: Acquire a first real-time vibration signal of a gas turbine rotor bearing in a first direction, a second real-time vibration signal in a second direction, and a real-time noise signal; wherein the first direction and the second direction are orthogonal;

[0082] Step 302: Generate a target RGB image based on the first real-time vibration signal, the second real-time vibration signal, and the real-time noise signal;

[0083] Step 303: Input the target RGB image into a trained vibration fault detection model to obtain the vibration fault type output by the vibration fault detection model; wherein the vibration fault detection model is trained based on sample RGB images and corresponding fault type labels.

[0084] The target RGB image and the sample RGB image are generated in the same way and will not be described again here.

[0085] In an embodiment of the present application, a first real-time vibration signal of a gas turbine rotor bearing in a first direction, a second real-time vibration signal in a second direction, and a real-time noise signal are obtained; wherein the first direction and the second direction are orthogonal; based on the first real-time vibration signal, the second real-time vibration signal and the real-time noise signal, a target RGB image is generated; the target RGB image is input into a trained vibration fault detection model to obtain the vibration fault type output by the vibration fault detection model; wherein the vibration fault detection model is trained based on sample RGB images and corresponding fault type labels. The present application converts the vibration signals and noise signals in two orthogonal directions into RGB images, which can effectively integrate the information of the vibration signal and the noise interference, enhance the image expression, enable the model to capture richer and more detailed features, and thus improve the accuracy of fault detection. In addition, the present application can generate RGB images in real time during the operation of the equipment and perform deep learning reasoning to timely predict the fault type of the equipment, thereby improving the preventive maintenance capability of the equipment, avoiding downtime losses caused by faults, and improving the reliability and operating efficiency of the equipment.

[0086] In order to implement the above embodiment, the embodiment of the present application also proposes a vibration fault detection model training device.

[0087] Figure 4 This is a structural diagram of a vibration fault detection model training device provided in an embodiment of the present application.

[0088] like Figure 4 As shown, the vibration fault detection model training device 400 includes:

[0089] An acquisition module 410 is configured to acquire a first vibration signal of a gas turbine rotor bearing in a first direction, a second vibration signal in a second direction, and a sample noise signal; the first direction and the second direction are orthogonal;

[0090] A generating module 420 is configured to generate a sample RGB image based on the first vibration signal, the second vibration signal, and the sample noise signal, and to obtain a fault type label corresponding to the sample RGB image;

[0091] The training module 430 is used to train the to-be-trained model based on the sample RGB images and the corresponding fault type labels to obtain a vibration fault detection model; wherein the vibration fault detection model is used to detect the vibration fault type of the gas turbine.

[0092] Optionally, the acquisition module 410 is specifically configured to:

[0093] Performing filtering and decomposition on the first vibration signal and the second vibration signal respectively to obtain a first noise signal in the first vibration signal and a second noise signal in the second vibration signal;

[0094] A sample noise signal is acquired based on the first noise signal and the second noise signal.

[0095] Optionally, the first noise signal and the second noise signal correspond to multiple moments, and the acquisition module 410 is specifically configured to:

[0096] For any target moment among the multiple moments, obtaining a first signal value of the first noise signal at the target moment and a second signal value of the second noise signal at the target moment;

[0097] Performing mean processing on the first signal value and the second signal value to obtain a mean signal value;

[0098] A sample noise signal is generated based on a mean signal value corresponding to each moment in a plurality of moments.

[0099] Optionally, the generating module 420 is specifically configured to:

[0100] According to the set image length, the first vibration signal, the second vibration signal, and the sample noise signal are respectively split to obtain corresponding multiple first vibration sub-signals, multiple second vibration sub-signals, and multiple target noise sub-signals;

[0101] In chronological order, the signal value corresponding to each first vibration sub-signal is used as a row of pixel values ​​in the first channel to obtain a first channel image; the signal value corresponding to each second vibration sub-signal is used as a row of pixel values ​​in the second channel to obtain a second channel image; the signal value corresponding to each target noise sub-signal is used as a row of pixel values ​​in the third channel to obtain a third channel image;

[0102] Generate a sample RGB image based on the first channel image, the second channel image, and the third channel image.

[0103] Optionally, the generating module 420 is specifically configured to:

[0104] According to the set image length, the first vibration signal is split to obtain a plurality of first sub-signals, the second vibration signal is split to obtain a plurality of second sub-signals, and the sample noise signal is split to obtain a plurality of third sub-signals;

[0105] For any target sub-signal among the first sub-signal, the second sub-signal, and the third sub-signal, normalizing the target sub-signal to obtain a corresponding normalized signal;

[0106] Based on the set mapping mode, the normalized signal is mapped to obtain a mapped signal; wherein the signal value in the mapped signal is within the set value range;

[0107] The mapped signal corresponding to the first sub-signal is used as the first vibration sub-signal, the mapped signal corresponding to the second sub-signal is used as the second vibration sub-signal, and the mapped signal corresponding to the third sub-signal is used as the target noise sub-signal.

[0108] Optionally, the generating module 420 is specifically configured to:

[0109] Use fast Fourier transform algorithm or wavelet transform algorithm to perform time-frequency analysis on sample RGB images;

[0110] Based on the time-frequency analysis results, the fault type label corresponding to the sample RGB image is determined.

[0111] It should be noted that the above explanation of the embodiment of the vibration fault detection model training method is also applicable to the vibration fault detection model training device of this embodiment, and will not be repeated here.

[0112] In order to implement the above embodiment, the embodiment of the present application further proposes a vibration fault detection device.

[0113] Figure 5 A schematic structural diagram of a vibration fault detection device provided in an embodiment of the present application.

[0114] like Figure 5 As shown, the vibration fault detection device 500 includes:

[0115] An acquisition module 510 is configured to acquire a first real-time vibration signal of a gas turbine rotor bearing in a first direction, a second real-time vibration signal in a second direction, and a real-time noise signal; wherein the first direction and the second direction are orthogonal;

[0116] A generating module 520 is configured to generate a target RGB image based on the first real-time vibration signal, the second real-time vibration signal, and the real-time noise signal;

[0117] The detection module 530 is used to input the target RGB image into a trained vibration fault detection model to obtain the vibration fault type output by the vibration fault detection model; wherein the vibration fault detection model is trained based on sample RGB images and corresponding fault type labels.

[0118] It should be noted that the above explanation of the embodiment of the vibration fault detection method is also applicable to the vibration fault detection device of this embodiment, and will not be repeated here.

[0119] Figure 6Schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Among them, the electronic device 600 in this embodiment is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0120] like Figure 6 As shown, the electronic device 600 includes:

[0121] The memory 601 and the processor 602, a bus 603 connecting different components (including the memory 601 and the processor 602), the memory 601 stores a computer program, and when the processor 602 executes the program, the vibration fault detection model training method or the vibration fault detection method of the embodiment of the present application is implemented.

[0122] Bus 603 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0123] The electronic device 600 typically includes a variety of electronic device readable media. These media can be any available media that can be accessed by the electronic device 600, including volatile and non-volatile media, removable and non-removable media.

[0124] The memory 601 may also include computer system readable media in the form of volatile memory, such as random access memory (RAM) 604 and / or cache memory 605. The electronic device 600 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 606 may be used to read and write non-removable, non-volatile magnetic media ( Figure 6 Not shown, often called a "hard drive"). Although Figure 6Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 603 via one or more data medium interfaces. Memory 601 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of various embodiments of the present application.

[0125] A program / utility 608 having a set (at least one) of program modules 607 may be stored, for example, in memory 601. Such program modules 607 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may include an implementation of a network environment. Program modules 607 generally implement the functions and / or methods of the embodiments described herein.

[0126] The electronic device 600 may also communicate with one or more external devices 609 (e.g., a keyboard, a pointing device, a display 611, etc.), and may also communicate with one or more devices that enable a user to interact with the electronic device 600, and / or any device that enables the electronic device 600 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication may be performed through an input / output (I / O) interface 612. Furthermore, the electronic device 600 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 613. Figure 6 As shown, the network adapter 613 communicates with other modules of the electronic device 600 via the bus 603. 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 600, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0127] The processor 602 executes various functional applications and data processing by running programs stored in the memory 601 .

[0128] It should be noted that the implementation process and technical principles of the electronic device of this embodiment can be found in the aforementioned explanation of the vibration fault detection model training method or the vibration fault detection method of the embodiment of the present application, and will not be repeated here.

[0129] In order to implement the above embodiments, the present application also proposes a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the methods provided by the above embodiments.

[0130] In order to implement the above embodiments, the present application also proposes a computer program product, including a computer program, which implements the methods provided by the above embodiments when executed by a processor.

[0131] The collection, storage, use, processing, transmission, provision and disclosure of user personal information involved in this application are in compliance with relevant laws and regulations and do not violate public order and good morals.

[0132] It is important to note that personal information collected from users should be used for legitimate and reasonable purposes and should not be shared or sold beyond these legitimate uses. Furthermore, such collection / sharing should be conducted only after receiving the user's informed consent, including but not limited to notifying the user to read the user agreement / user notice and sign an agreement / authorization that includes the relevant user information before using the feature. Furthermore, any necessary steps must be taken to safeguard and secure access to such personal information and ensure that others with access to personal information comply with its privacy policy and procedures.

[0133] This application contemplates providing implementations that allow users to selectively block the use or access of personal information data. Specifically, this disclosure contemplates providing hardware and / or software to prevent or block access to such personal information data. Risks can be minimized by limiting data collection and deleting data once it is no longer needed. Furthermore, where applicable, such personal information can be de-identified to protect user privacy.

[0134] In the descriptions of the foregoing embodiments, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are mutually inconsistent.

[0135] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0136] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0137] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing it in another suitable manner if necessary, and then storing it in a computer memory.

[0138] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0139] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0140] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0141] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A vibration fault detection model training method, characterized in that: The following steps are involved: Acquire a first vibration signal of a gas turbine rotor bearing in a first direction, a second vibration signal in a second direction, and a sample noise signal; the first direction and the second direction are orthogonal; generating a sample RGB image based on the first vibration signal, the second vibration signal, and the sample noise signal, and obtaining a fault type label corresponding to the sample RGB image; Based on the sample RGB images and the corresponding fault type labels, the to-be-trained model is trained to obtain a vibration fault detection model; wherein the vibration fault detection model is used to detect the vibration fault type of the gas turbine.

2. The method according to claim 1, characterized in that Acquiring the sample noise signal includes: performing filtering and decomposition on the first vibration signal and the second vibration signal respectively to obtain a first noise signal in the first vibration signal and a second noise signal in the second vibration signal; The sample noise signal is acquired based on the first noise signal and the second noise signal.

3. The method according to claim 2, characterized in that The first noise signal and the second noise signal correspond to a plurality of time instants, and obtaining the sample noise signal based on the first noise signal and the second noise signal includes: For any target moment among the multiple moments, obtaining a first signal value at the target moment in the first noise signal and a second signal value at the target moment in the second noise signal; performing mean processing on the first signal value and the second signal value to obtain a mean signal value; The sample noise signal is generated based on the mean signal value corresponding to each moment in the multiple moments.

4. The method according to claim 1, wherein The generating of a sample RGB image based on the first vibration signal, the second vibration signal, and the sample noise signal includes: Splitting the first vibration signal, the second vibration signal, and the sample noise signal according to a set image length to obtain corresponding multiple first vibration sub-signals, multiple second vibration sub-signals, and multiple target noise sub-signals; In chronological order, the signal values ​​corresponding to each of the first vibration sub-signals are used as pixel values ​​for a row in a first channel to obtain a first channel image; the signal values ​​corresponding to each of the second vibration sub-signals are used as pixel values ​​for a row in a second channel to obtain a second channel image; and the signal values ​​corresponding to each of the target noise sub-signals are used as pixel values ​​for a row in a third channel to obtain a third channel image. The sample RGB image is generated based on the first channel image, the second channel image, and the third channel image.

5. The method according to claim 4, characterized in that The first vibration signal, the second vibration signal, and the sample noise signal are respectively split according to the set image length to obtain corresponding multiple first vibration sub-signals, multiple second vibration sub-signals, and multiple target noise sub-signals, including: According to a set image length, the first vibration signal is split to obtain a plurality of first sub-signals, the second vibration signal is split to obtain a plurality of second sub-signals, and the sample noise signal is split to obtain a plurality of third sub-signals; For any target sub-signal among the first sub-signal, the second sub-signal, and the third sub-signal, performing normalization processing on the target sub-signal to obtain a corresponding normalized signal; Based on the set mapping mode, mapping processing is performed on the normalized signal to obtain a mapped signal; wherein the signal value in the mapped signal is within the set value range; The mapped signal corresponding to the first sub-signal is used as the first vibration sub-signal, the mapped signal corresponding to the second sub-signal is used as the second vibration sub-signal, and the mapped signal corresponding to the third sub-signal is used as the target noise sub-signal.

6. The method according to claim 1, characterized in that The obtaining of the fault type label corresponding to the sample RGB image includes: Performing time-frequency analysis on the sample RGB image using a fast Fourier transform algorithm or a wavelet transform algorithm; Based on the time-frequency analysis processing result, the fault type label corresponding to the sample RGB image is determined.

7. A vibration fault detection method, characterized in that: The following steps are involved: Acquire a first real-time vibration signal of a gas turbine rotor bearing in a first direction, a second real-time vibration signal in a second direction, and a real-time noise signal; wherein the first direction and the second direction are orthogonal; generating a target RGB image based on the first real-time vibration signal, the second real-time vibration signal, and the real-time noise signal; The target RGB image is input into a trained vibration fault detection model to obtain the vibration fault type output by the vibration fault detection model; wherein the vibration fault detection model is trained based on sample RGB images and corresponding fault type labels.

8. A vibration fault detection model training device, characterized in that: include: an acquisition module, configured to acquire a first vibration signal of a gas turbine rotor bearing in a first direction, a second vibration signal in a second direction, and a sample noise signal; the first direction and the second direction being orthogonal; a generating module, configured to generate a sample RGB image based on the first vibration signal, the second vibration signal, and the sample noise signal, and to obtain a fault type label corresponding to the sample RGB image; A training module is used to train a to-be-trained model based on the sample RGB image and the corresponding fault type label to obtain a vibration fault detection model; wherein the vibration fault detection model is used to detect the vibration fault type of the gas turbine.

9. A vibration fault detection device, characterized in that: include: an acquisition module, configured to acquire a first real-time vibration signal of a gas turbine rotor bearing in a first direction, a second real-time vibration signal in a second direction, and a real-time noise signal; wherein the first direction and the second direction are orthogonal; a generating module, configured to generate a target RGB image based on the first real-time vibration signal, the second real-time vibration signal, and the real-time noise signal; The detection module is used to input the target RGB image into a trained vibration fault detection model to obtain the vibration fault type output by the vibration fault detection model; wherein the vibration fault detection model is trained based on sample RGB images and corresponding fault type labels.

10. An electronic device, characterized in that: include: a processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 6 or the method according to claim 7.

Images