Method, device and medium for predicting the axis trajectory of a rotating device

By filtering and clustering the vibration data of the rotating device, an axial trajectory image recognition network is constructed, which solves the problem that traditional methods fail to consider the different working conditions of the axial trajectory, and realizes robust recognition and prediction of the axial trajectory , improves the automation and accuracy of fault diagnosis.

CN117593554BActive Publication Date: 2024-12-06KSB SHANGHAI PUMP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202311834030.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-12-28
Publication Date
2024-12-06
Estimated Expiration
2043-12-28

AI Technical Summary

Technical Problem

Traditional methods fail to effectively consider the shape characteristics of different working conditions such as translation, expansion, rotation and reflection of the axis trajectory when predicting the axis trajectory of the rotating equipment, resulting in a low degree of automation diagnosis and inaccurate fault identification.

Method used

By acquiring the vibration data of the rotating device, filtering and clustering, an axial trajectory image recognition network is constructed, and the shape changes of the axial trajectory are monitored online using deep learning methods to predict the type and probability of the axial trajectory.

Benefits of technology

It realizes robust identification and prediction of the axis trajectory of rotating equipment, improves the degree of automation and accuracy of fault diagnosis, reduces equipment costs, and is suitable for online fault monitoring.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117593554B_ABST
    Figure CN117593554B_ABST
Patent Text Reader

Abstract

An embodiment of the present invention relates to a method, device and medium for predicting the axis trajectory of a rotating device, including: acquiring vibration data of a rotating device for multiple periods within a certain frequency range; performing processing on the acquired vibration data to obtain an axis trajectory image corresponding to the processed vibration data; dividing the vibration data of the axis trajectory image into multiple separate samples according to the period, thereby clustering the multiple separate samples to obtain a clustered axis trajectory image; constructing an axis trajectory image recognition network based on the clustered axis trajectory images; and predicting the axis trajectory of the rotating device based on the constructed axis trajectory image recognition network.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of mechanical fault diagnosis and industrial control, and more particularly to a method, a computing device and a storage medium for predicting the axis trajectory of a rotating device. Background Art

[0002] Traditional methods for detecting faults in rotating equipment (such as but not limited to rotating mechanical equipment such as pumps and motors) are, for example, based on mechanism models, and fault judgment is performed by detecting whether there is a peak in the spectrum at a specific frequency. For example, the mechanism model mainly uses mechanism rules to determine whether a fault has occurred.

[0003] In traditional diagnostic systems, the axis trajectory is manually identified, which greatly affects the degree of automation of diagnosis. Commonly used automatic identification methods mainly include feature extraction and classification recognition. Among them, feature extraction is generally divided into two categories: one is the extraction of transient signal features, and the other is shape analysis of image processing (i.e., various transformations are made to the exterior or shape of the region to extract the shape features of the region, such as the moment method); classification recognition methods are mainly machine learning and deep learning, including some support vector machines, or image recognition methods.

[0004] Conventional methods believe that the recognition of axis trajectory is essentially a problem of invariant pattern recognition of two-dimensional graphics, that is, the result of pattern recognition does not change with the translation, expansion, rotation or even reflection of the graphics. However, in reality, the axis trajectory will produce the above changes.

[0005] In summary, the traditional method for predicting the axis trajectory of rotating equipment has the following shortcomings: there is a lack of axis trajectory prediction that takes into account the shape characteristics of the axis trajectory under different working conditions such as translation, extension, rotation and reflection. Summary of the invention

[0006] In view of the above problems, the present invention provides a method, a computing device and a storage medium for predicting the axis trajectory of a rotating device, which can calculate the probability of occurrence of various faults of the rotating device.

[0007] According to a first aspect of the present invention, a method for predicting the axis trajectory of a rotating device is provided, comprising: acquiring vibration data (vibration acceleration data, vibration velocity data) of a rotating device for multiple periods within a certain frequency range; performing processing on the acquired vibration data to obtain an axis trajectory image corresponding to the processed vibration data; dividing the vibration data of the axis trajectory image into multiple individual samples according to the period, thereby clustering the multiple individual samples to obtain clustered vibration data and their corresponding axis trajectory images; constructing an axis trajectory image recognition network based on the clustered axis trajectory images; and predicting the axis trajectory of the rotating device based on the constructed axis trajectory image recognition network.

[0008] In one embodiment, obtaining vibration data of multiple cycles of a rotating device within a certain frequency range includes: obtaining three-axis waveform data of the rotating device in the horizontal direction, vertical direction, and axial direction; determining the frequency to be extracted of the axis trajectory; and performing filtering processing on the three-axis waveform data based on the determined frequency to be extracted, thereby obtaining vibration data of multiple cycles corresponding to the frequency of the filtering processing performed.

[0009] In one embodiment, the filtering process includes one or more of the following processes: low-pass filtering, high-pass filtering, band-pass filtering, and Fourier transform.

[0010] In one embodiment, performing processing on the acquired axis trajectory includes: acquiring the determination requirement of the axis trajectory; selecting corresponding axial data based on the acquired determination requirement; and drawing the selected axial data in a polar coordinate system, thereby achieving drawing of the axis trajectory.

[0011] In one embodiment, clustering the multiple individual samples includes: determining whether the axial trajectory image includes more than two periodic shape patterns; in response to determining that the axial trajectory image does not include more than two periodic shape patterns, determining the axial trajectory image as a clustered axial trajectory image; and in response to determining that the axial trajectory image includes more than two periodic shape patterns, separately determining multiple periodic shape patterns of the axial trajectory image; performing data weighted merging on the determined two or more periodic shape patterns to determine a clustered axial trajectory image based on the weighted merged periodic shape patterns.

[0012] In one embodiment, determining whether the axis trajectory image includes more than two periodic shape patterns includes: determining whether the axis trajectory image includes more than two periodic shape patterns based on partitioning clustering, density clustering or hierarchical clustering.

[0013] In one embodiment, the input of the axis track image recognition network is the clustered axis track image, and the output is the recognition result and recognition prediction probability of the axis track image.

[0014] In one embodiment, predicting the axis trajectory of the rotating device includes: merging and splicing a plurality of clustered axis trajectory images corresponding to a predetermined time period to obtain a continuous axis trajectory image corresponding to the predetermined time period; and determining the axis trajectory type and prediction probability of the rotating device in a future cycle based on an axis trajectory image recognition network and the continuous axis trajectory image.

[0015] In one embodiment, the axis trajectory includes a primary type and a secondary type corresponding to the primary type, the primary type includes at least a banana shape, an ellipse, an inner eight-shaped shape, a multiple eight-shaped shape and an outer eight-shaped shape, and the secondary type includes at least a solid circle, a hollow circle, a horizontal ellipse, a vertical ellipse, a flat ellipse biased to the left, a normal ellipse biased to the left, a flat ellipse biased to the right, and a normal ellipse biased to the right.

[0016] According to a second aspect of the present invention, a computing device is provided, which comprises: at least one processing unit; at least one memory, which is coupled to the at least one processing unit and stores instructions for execution by the at least one processing unit, and when the instructions are executed by the at least one processing unit, the device performs the steps of the method according to the first aspect.

[0017] According to a third aspect of the present invention, there is provided a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a machine, the method according to the first aspect is implemented.

[0018] Using the above technical means, the vibration data can be Fourier transformed and bandpass filtered based on expert experience, and clustered to classify and merge vibration data of multiple cycles to construct the axis trajectory image. Then, based on the deep learning method, a neural network is used to build a recognition model to continuously monitor the shape of the axis trajectory online and predict the image changes of the axis trajectory.

[0019] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other features, advantages and aspects of the embodiments of the present invention will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. In the accompanying drawings, the same or similar reference numerals represent the same or similar elements.

[0021] Figure 1 A schematic diagram of a system for implementing a method for predicting the axis trajectory of a rotating device according to an embodiment of the present invention is shown.

[0022] Figure 2 A flow chart of a method 200 for predicting the axis trajectory of a rotating device according to an embodiment of the present invention is shown.

[0023] Figure 3 A schematic diagram of a purified multi-period axis trajectory according to an embodiment of the present invention is shown.

[0024] Figure 4 A schematic diagram of axis trajectory prediction according to an embodiment of the present invention is shown.

[0025] Figure 5 A schematic diagram of an example electronic device 500 is shown that can be used to implement embodiments of the present disclosure. DETAILED DESCRIPTION

[0026] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0027] As used herein, the term "including" and its variations mean open inclusion, i.e., "including but not limited to". Unless otherwise stated, the term "or" means "and / or". The term "based on" means "based at least in part on". The terms "an example embodiment" and "an embodiment" mean "at least one example embodiment". The term "another embodiment" means "at least one additional embodiment". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0028] As described above, the conventional method for predicting the axis trajectory of a rotating device does not take into account the change of the axis trajectory within multiple cycles. In order to at least partially solve the above problems and one or more of other potential problems, an exemplary embodiment of the present invention proposes a solution for predicting the axis trajectory of a rotating device.

[0029] In traditional methods, the identification of axis trajectory is generally only for the single-cycle axis trajectory obtained after signal decomposition and reconstruction, and the different working conditions of axis trajectory such as translation, extension, rotation, etc. are not distinguished in the identification process. There is a lack of professional engineering knowledge, and the experimental process is mostly limited to simulation. There is little engineering practice, which makes it difficult to provide a reliable basis for subsequent fault identification. In the actual field, test engineers use a variety of professional vibration analysis equipment to draw axis trajectory. The drawn images often contain axis trajectory of multiple cycles and multiple frequencies over a period of time. After that, the axis trajectory shape is identified through manual experience, and then subsequent fault identification is carried out. The cost of such equipment is often very high. Historical axis trajectory analysis is limited to the identification of axis trajectory, and does not involve online prediction of axis trajectory. The present invention provides an online axis trajectory drawing method that takes into account both theoretical and practical engineering needs. After decomposing and reconstructing the signal, a multi-period axis trajectory that conforms to actual engineering implementation is reconstructed. Combined with the existing mature image recognition technology, the shape characteristics of the axis trajectory under different working conditions such as translation, expansion, rotation and reflection are fully considered to provide a robust axis trajectory recognition and prediction method.

[0030] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments, and they cannot be understood as limiting the scope of protection of this application. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0031] Figure 1 FIG. 2 is a schematic diagram showing a system for implementing a method for predicting the axis trajectory of a rotating device according to an embodiment of the present invention. Figure 1 As shown in , the system includes a computing device 110, a rotating device 130, and an acceleration detection device 140. In some embodiments, the computing device 110, the rotating device 130, and the acceleration detection device 140 are connected in communication and perform data exchange directly or via a network.

[0032] The rotating device 130 may be, for example but not limited to, a rotating device driven by a variable frequency or a fixed frequency, such as a pump, a water pump, etc.

[0033] Regarding the computing device 110, it is used to calculate the failure probability of the rotating device 130. Specifically, the computing device 110 is used to obtain vibration data of multiple cycles of the rotating device within a certain frequency range; perform processing on the obtained vibration data to obtain an axis trajectory image corresponding to the processed vibration data; divide the vibration data of the axis trajectory image into multiple individual samples according to the period, and perform clustering on the multiple individual samples to obtain clustered vibration data and its corresponding axis trajectory map; construct an axis trajectory image recognition network based on the clustered axis trajectory image; and predict the axis trajectory of the rotating device based on the constructed axis trajectory image recognition network.

[0034] In some embodiments, the computing device 110 may have one or more processing units, including dedicated processing units such as GPU, FPGA and ASIC, and general processing units such as CPU. In addition, one or more virtual machines may also be running on each computing device 110. The computing device 110 may include an acquisition module 112, a processing module 114, a clustering module 116, an identification module 118, and a prediction module 120.

[0035] Figure 2 FIG. 2 is a flow chart of a method 200 for predicting the axis trajectory of a rotating device according to an embodiment of the present invention. The method 200 may be performed as follows: Figure 1 The computing device 110 shown may also be executed in Figure 5 The method 200 is executed at the electronic device 500. It should be understood that the method 200 may further include additional steps not shown and / or may omit the steps shown, and the scope of the present invention is not limited in this respect.

[0036] In step 202, the computing device 110 obtains vibration data of a rotating device over a certain frequency range for multiple cycles. The rotating device may be, for example but not limited to, a fixed frequency or variable frequency driven pump, motor, etc.

[0037] In one embodiment, the computing device 110 can obtain the axis trajectory data of the rotating device. This is usually done by using an accelerometer, a gyroscope or other sensors. These sensors will provide three-axis waveform data in the horizontal direction, vertical direction and axial direction. For specific details of the parameter values. Taking sensors as an example, MEMS accelerometers and gyroscopes can be used. Sampling frequency: set to 1000 Hz to ensure high-frequency data acquisition. Data acquisition time: In order to obtain data for multiple cycles, the acquisition time should be set according to the required number of cycles, such as 5 seconds, so as to obtain the three-axis waveform data in the horizontal direction, vertical direction and axial direction of the rotating device.

[0038] The frequency range may be, for example, 0.36-100 Hz, see Figure 3 .

[0039] The computing device 110 may perform filtering processing on the acquired data, and the processing specifically includes performing low-pass filtering, high-pass filtering, band-pass filtering, and Fourier transform processing to obtain a frequency domain representation of the vibration data. The frequency to be extracted may be determined according to specific application requirements. For example, a frequency to be extracted of 10 Hz is selected for analyzing the axis trajectory. Low-pass filtering may use a Butterworth low-pass filter. The cutoff frequency is set to half the frequency to be extracted, i.e., 5 Hz. The filter order selects the 4th order to smooth the data. High-pass filtering may use a Butterworth high-pass filter. The cutoff frequency is set to 0.5 Hz to remove low-frequency noise. Filter order: Select the 4th order to smooth the data. Band-pass filtering may use a Butterworth band-pass filter. The low-pass cutoff frequency is set to 5 Hz. The high-pass cutoff frequency is set to 0.5 Hz. The filter order selects the 4th order to retain the signal within the target frequency range. Perform a discrete Fourier transform (DFT) to convert the time domain data into frequency domain data. Parameters of discrete Fourier transform: sampling frequency, signal length, etc.

[0040] Based on the collected waveform data in three directions: horizontal, vertical, and axial, vibration data of multiple cycles within a predetermined frequency range can be extracted, such as the axis trajectory, with an accuracy that is better than the purification level that can be provided by general vibration meters.

[0041] At step 204 , the computing device 110 performs processing on the acquired vibration data to acquire an axis trajectory image corresponding to the processed vibration data.

[0042] In one embodiment, the computing device 110 obtains the identification requirement of the axis trajectory; selects corresponding axial data based on the obtained identification requirement; and plots the selected axial data in a polar coordinate system, thereby completing the preliminary plotting of the axis trajectory.

[0043] The computing device 110 selects the required axial data according to the input determination requirements of the axis trajectory. For example, in general, only horizontal data and vertical data are selected. In special cases, axial data is added and the data is plotted in a polar coordinate system, thus completing the preliminary drawing of the axis trajectory.

[0044] Figure 3 FIG. 2 shows a schematic diagram of a purified multi-period axis trajectory according to an embodiment of the present invention. Figure 3As shown in the figure, the original reference axis trajectory including multiple cycles is relatively messy and cannot be parsed and identified. However, after correlation processing and reconstruction in the frequency domain, by extracting vibration data of multiple cycles in a fixed frequency, such as the axis trajectory, the accuracy is better than the purification level that can be provided by a general vibration meter; at the same time, the clustering reconstruction of the axis trajectory, compared with the single-cycle axis trajectory recognition, better reflects the stage-by-stage characteristics of the axis trajectory rather than focusing on the instantaneous state of the axis trajectory, which is more suitable for actual engineering applications.

[0045] In step 206 , the computing device 110 divides the vibration data of the shaft center trajectory image into a plurality of individual samples according to a period, and performs clustering on the plurality of individual samples to obtain clustered vibration data and its corresponding shaft center trajectory image.

[0046] In one embodiment, the computing device 110 may divide the image into multiple samples according to the period and determine the size of the division window. The clustering algorithm may select, for example, K-means clustering, hierarchical clustering, etc., and set the number of clusters. Feature extraction includes extracting features of each sample, such as frequency domain components, peak information, etc., so as to determine whether the vibration data includes more than two periodic shape patterns.

[0047] If the vibration data does not include more than two periodic shape patterns, the axis trajectory image is determined as a clustered axis trajectory image. This means that the image already contains a major periodic shape. If the vibration data includes more than two periodic shape patterns, the following operations are performed: multiple periodic shape patterns of the axis trajectory image are determined separately. Data weighted merging is performed on the determined more than two periodic shape patterns to determine a clustered axis trajectory image based on the weighted merged periodic shape patterns. For determining whether the axis trajectory image includes more than two periodic shape patterns, partitioning clustering, density-based clustering or hierarchical clustering can be used. These algorithms will help identify axis trajectories of different periodic shapes.

[0048] For example, considering that the axis trajectory image constructed after data purification may still contain multiple shape patterns with large deviations in directions and angles, the purified axis trajectory data is divided into a separate sample data according to a single cycle, and these sample data are clustered to determine whether the axis trajectory data used contains more than one periodic shape pattern with a large proportion: if the axis trajectory data does not contain a variety of different main periodic shape patterns, there is no need to reconstruct the original axis trajectory. If the axis trajectory data contains a variety of different main shape patterns, the single-cycle data of the axis trajectory is weighted and adjusted according to the proportion of the center data of the main shape pattern of the axis trajectory to the main shape pattern, and the adjusted new data is used to reconstruct the axis trajectory image.

[0049] An example of weighted merging adjustment is as follows: If there are two main periodic shape modes A and B, the central data of these two periodic shape modes are Core_A and Core_B, and the proportions are weight_A and weight_B respectively. The single-period data before adjustment is Data, which belongs to shape mode A. Then the single-period data after adjustment can be expressed as

[0050] Adjust_Data=Core_A×weight_A+Core_B×weight_B.

[0051] In step 208 , the computing device 110 constructs an axis track image recognition network based on the clustered axis track images.

[0052] In one embodiment, the computing device 110 can simulate the axis trajectory image and the corresponding shape label according to the actual data to form an axis trajectory shape sample library. Under the condition that the axis trajectory image is not randomly stretched, rotated or flipped, the axis trajectory image in the sample library is data enhanced, and then an axis trajectory image recognition network is constructed based on a neural network.

[0053] The computing device 110 can completely capture the image according to the center and perform data normalization processing as the model input; the model output is the image recognition result and recognition prediction probability; the neural network model used is not limited to type, and the appropriate network model is selected for training and prediction based on the size, complexity, and number of image samples of the axis trajectory image.

[0054] Regarding the neural network models used for training, for example, the Visual Geometry Group (VGG) model of Oxford University, the Residual Network (RNN)

[0055] Network, ResNet) model, Convolutional Neural Network

[0056] Network, CNN), fully connected neural network (FCN), generative adversarial network (GAN), recurrent neural network (RNN), long short-term memory network (LSTM), etc. The typical model training method can use the script-style Pytorch training method, taking into account both CPU model training and GPU model training.

[0057] While the axis trajectory is being recognized online, the axis trajectory images and labels recognized online are added to the axis trajectory shape sample library to support subsequent iterations of the axis trajectory image recognition network.

[0058] The axis track image recognition network inputs the clustered axis track image and outputs the recognition result and recognition prediction probability of the axis track image.

[0059] In step 210 , the computing device 110 predicts the axis trajectory of the rotating device based on the constructed axis trajectory image recognition network.

[0060] In one embodiment, the computing device 110 can collect vibration data of the equipment in real time during operation, identify the axis trajectory based on the axis trajectory image recognition network constructed in step 208, continuously monitor the shape of the axis trajectory online, and predict the image changes of the axis trajectory. For example, the continuous historical axis trajectory images within a period of time are merged and spliced, and annotated as the input of the axis trajectory image recognition network. Under the condition that these axis trajectory images are not randomly stretched, rotated, or flipped, the image prediction of the axis trajectory is performed, and the output result is the predicted axis trajectory type and the corresponding prediction probability. According to the monitoring and prediction of the above-mentioned axis trajectory, combined with other indicators of the equipment, comprehensive consideration is taken to assist in subsequent fault identification.

[0061] Figure 4 FIG. 2 shows a schematic diagram of axis trajectory prediction according to an embodiment of the present invention. Figure 4 As shown, the axis trajectory involved in the present invention includes a primary type and a secondary type corresponding to the primary type, wherein the primary type includes at least a banana shape, an ellipse, an inner eight-shaped shape, a multi-eight-shaped shape and an outer eight-shaped shape, and the secondary type includes at least a solid circle, a hollow circle, a horizontal ellipse, a vertical ellipse, a left-biased flat ellipse, a left-biased normal ellipse, a right-biased flat ellipse, and a right-biased normal ellipse. The computing device 110 can merge and splice a plurality of clustered axis trajectory images corresponding to a predetermined time period, thereby obtaining a continuous axis trajectory image corresponding to the predetermined time period; and determine the axis trajectory type and prediction probability of the future cycle of the rotating device based on the axis trajectory image recognition network and the continuous axis trajectory image.

[0062] Using the above technical means, the vibration data can be Fourier transformed and bandpass filtered based on expert experience, and clustered to classify and merge multiple cycles of vibration data, such as axis trajectory, to construct an axis trajectory image. Then, based on the deep learning method, a neural network is used to construct an axis trajectory image recognition network, which continuously monitors the shape of the axis trajectory online and predicts the image changes of the axis trajectory.

[0063] In terms of axis trajectory drawing, the present invention can extract vibration data of multiple cycles in a fixed frequency, such as the axis trajectory, with an accuracy that is better than the purification level that can be provided by a general vibration meter; at the same time, compared with the single-cycle axis trajectory recognition, it better reflects the stage-by-stage characteristic performance of the axis trajectory rather than focusing on the instantaneous state of the axis trajectory, which is more in line with actual engineering applications.

[0064] In terms of axis trajectory recognition, the present invention can realize automatic axis trajectory recognition without relying on complex feature engineering, and fully consider the shape characteristics of the axis trajectory under different working conditions such as translation, extension, rotation and reflection, and at the same time give the recognition prediction probability of the axis trajectory; in addition, the axis trajectory shape sample library can be expanded according to the real-time data situation, and the axis trajectory image recognition network can be continuously optimized; finally, in the use of the axis trajectory recognition results, axis trajectory monitoring, axis trajectory prediction and other data indicators are used to conduct comprehensive analysis to assist fault judgment.

[0065] Figure 5 A schematic diagram of an example electronic device 500 that can be used to implement an embodiment of the present specification is shown. As shown, the electronic device 500 includes a central processing unit (CPU) 501, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 502 or computer program instructions loaded from a storage unit 508 into a random access memory (RAM) 503. In the random access memory 503, various programs and data required for the operation of the electronic device 500 can also be stored. The central processing unit 501, the read-only memory 502, and the random access memory 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0066] Multiple components in the electronic device 500 are connected to the input / output interface 505, including: an input unit 506, such as a keyboard, a mouse, a microphone, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the electronic device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0067] The various processes and processing described above, such as method 200, may be performed by the central processing unit 501. For example, in some embodiments, method 200 may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via the read-only memory 502 and / or the communication unit 509. When the computer program is loaded into the random access memory 503 and executed by the central processing unit 501, one or more actions of the method 200 described above may be performed.

[0068] The present invention relates to methods, apparatuses, systems, electronic devices, computer-readable storage media and / or computer program products. The computer program products may include computer-readable program instructions for executing various aspects of the present invention.

[0069] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: 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), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0070] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network may include copper transmission cables, optical fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge computing devices. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in each computing / processing device.

[0071] The computer program instructions for performing the operation of the present invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. Computer-readable program instructions may be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect via the Internet). In some embodiments, an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), may be personalized by utilizing the state information of the computer-readable program instructions, and the electronic circuit may execute the computer-readable program instructions, thereby realizing various aspects of the present invention.

[0072] Various aspects of the present invention are described herein with reference to the flowcharts and / or schematic diagrams of the methods, devices (systems) and computer program products according to embodiments of the present invention. It should be understood that each step of the flowchart and / or schematic diagram and the combination of each step in the flowchart and / or schematic diagram can be implemented by computer-readable program instructions.

[0073] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more steps in the flowchart and / or schematic diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions make the computer, programmable data processing device, and / or other equipment work in a specific way, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more steps in the flowchart and / or schematic diagram.

[0074] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operational steps are executed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more steps in the flowchart and / or schematic diagram.

[0075] The flowcharts and schematic diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to multiple embodiments of the present invention. In this regard, each step in the flowchart or schematic diagram can represent a module, a program segment or a part of an instruction, and a module, a program segment or a part of an instruction contains one or more executable instructions for realizing a specified logical function. In some alternative implementations, the functions marked in the steps can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive steps can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each step in the schematic diagram and / or flowchart, and the combination of multiple steps in the schematic diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0076] The embodiments of the present invention have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for predicting the axis trajectory of a rotating device, comprising: Obtain vibration data of rotating equipment for multiple cycles within a certain frequency range; Processing the acquired vibration data to acquire an axis center trajectory image corresponding to the processed vibration data; Dividing the vibration data of the axis center trajectory image into a plurality of individual samples according to a period, thereby performing clustering on the plurality of individual samples to obtain clustered vibration data and the axis center trajectory image corresponding thereto; Based on the clustered axis track images, construct an axis track image recognition network, wherein the input of the axis track image recognition network is the clustered axis track image, and the output is the recognition result and recognition prediction probability of the axis track image; and Based on the constructed axis trajectory image recognition network, the axis trajectory of the rotating equipment is predicted, and multiple clustered axis trajectory images corresponding to the predetermined time period are merged and spliced ​​to obtain a continuous axis trajectory image corresponding to the predetermined time period. Based on the axis trajectory image recognition network and the continuous axis trajectory images, the axis trajectory type and prediction probability of the future cycle of the rotating equipment are determined.

2. The method according to claim 1, characterized in that Obtaining vibration data of rotating equipment for multiple cycles within a certain frequency range includes: Acquire three-axis waveform data of the rotating device in the horizontal direction, the vertical direction, and the axial direction; determining the frequency of the axis trajectory to be extracted; and A filter process is performed on the triaxial waveform data based on the determined frequency to be extracted, thereby acquiring vibration data of a plurality of cycles corresponding to the frequency of the performed filter process.

3. The method according to claim 2, characterized in that The filtering process includes one or more of the following processes: low-pass filtering, high-pass filtering, band-pass filtering, and Fourier transform.

4. The method according to claim 1, characterized in that The processing performed on the acquired axis trajectory includes: Obtain the determination requirements of the axis trajectory; Selecting corresponding axial data based on the obtained discrimination requirements; and The selected axial data is plotted in the polar coordinate system to achieve the drawing of the axis trajectory.

5. The method according to claim 1, characterized in that Performing clustering on the plurality of individual samples comprises: determining whether the axis trajectory image includes two or more periodic shape patterns; In response to determining that the axis-track image does not include two or more periodic shape patterns, determining the axis-track image as a clustered axis-track image; and In response to determining that the axis trajectory image includes more than two periodic shape modes, respectively determining a plurality of periodic shape modes of the axis trajectory image; A data weighted combination is performed on the determined two or more periodic shape patterns to determine a clustered axis trajectory image based on the weighted combined periodic shape patterns.

6. The method according to claim 5, characterized in that Determining whether the axis trajectory image includes two or more periodic shape patterns includes: It is determined whether the axis trajectory image includes more than two periodic shape patterns based on a partitioning clustering method, a density-based clustering method, or a hierarchical clustering method.

7. The method according to any one of claims 1 to 6, characterized in that: The axis trajectory includes a primary type and a secondary type corresponding to the primary type, the primary type includes at least a banana shape, an ellipse, an inner eight-shaped shape, a multiple eight-shaped shape and an outer eight-shaped shape, and the secondary type includes at least a solid circle, a hollow circle, a horizontal ellipse, a vertical ellipse, a flat ellipse biased to the left, a normal ellipse biased to the left, a flat ellipse biased to the right, and a normal ellipse biased to the right.

8. A computing device comprising: at least one processing unit; At least one memory, the at least one memory being coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the computing device to perform the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method according to any one of claims 1 to 7 when executed by a machine.