Rotating equipment fault diagnosis method and device

By collecting and analyzing the weight distribution, installation records, operation images and vibration signals of the rotating equipment, identifying and extracting the fault deviation distribution of the rotating equipment, the problem of inaccurate fault detection in the prior art is solved, and the deviation failure of the rotating equipment is achieved Accurate diagnosis.

CN118708914BActive Publication Date: 2024-11-05LEVI INTELLIGENT (SHENZHEN) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411194933.9
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-11-05
Estimated Expiration
2044-08-29

AI Technical Summary

Technical Problem

The prior art fails to consider the operating characteristics of the equipment in the deviation fault diagnosis of rotating equipment, resulting in inaccurate fault detection.

Method used

By collecting weight distribution information and installation record information of the rotating equipment, combining the running image sequence and vibration signal sequence, rotation deviation recognition and aging deviation analysis are performed, and the fault deviation distribution is extracted to identify the fault diagnosis results.

Benefits of technology

It improves the accuracy of fault diagnosis of rotating equipment, can reflect the actual deviation of the equipment, eliminates the impact of aging deviation caused by the equipment's own characteristics, and improves the accuracy of fault monitoring.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118708914B_ABST
    Figure CN118708914B_ABST
Patent Text Reader

Abstract

The present invention relates to a method and device for diagnosing faults of rotating equipment, and relates to the field of data processing technology, comprising: collecting equipment feature information of a rotating equipment to be fault diagnosed, wherein the equipment feature information includes weight distribution information and installation record information of the rotating equipment; during the rotating operation of the rotating equipment, collecting an operation image sequence and a vibration signal sequence of the rotating equipment, performing rotation deviation identification according to the operation image sequence, and obtaining a basic deviation distribution; performing rotation deviation identification according to the vibration signal sequence, adjusting the basic deviation distribution, and obtaining a rotation deviation distribution; performing aging deviation analysis of the rotating equipment according to the equipment feature information, and obtaining an aging deviation distribution, performing aging deviation extraction within the rotation deviation distribution, and obtaining a fault deviation distribution, and identifying and obtaining a fault diagnosis result of the rotating equipment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and device for diagnosing faults of rotating equipment. Background Art

[0002] Rotating equipment will experience deviation faults from the preset rotation center when under rotating load. At present, the deviation faults of rotating equipment are generally detected in a single or fixed period, and directly judge whether a deviation fault occurs without considering the specific operating characteristics of each rotating equipment. Such fault diagnosis logic often cannot reflect the actual deviation fault situation of the equipment, resulting in inaccurate fault detection and diagnosis. Summary of the invention

[0003] The present invention aims to solve the technical problem that the deviation fault diagnosis of rotating equipment in the prior art does not take the equipment operation characteristics into consideration and cannot reflect the actual deviation fault situation of the equipment, resulting in inaccurate deviation fault detection and diagnosis.

[0004] In a first aspect, the present invention provides a method for diagnosing a fault of a rotating device, comprising: collecting device characteristic information of a rotating device to be diagnosed for a fault, wherein the device characteristic information includes weight distribution information and installation record information of the rotating device;

[0005] During the rotation operation of the rotating device, an operation image sequence and a vibration signal sequence of the rotating device are collected, and rotation deviation is identified according to the operation image sequence to obtain a basic deviation distribution;

[0006] Performing rotation deviation identification according to the vibration signal sequence, adjusting the basic deviation distribution, and obtaining a rotation deviation distribution;

[0007] According to the equipment characteristic information, an aging deviation analysis of the rotating equipment is performed to obtain an aging deviation distribution, an aging deviation extraction is performed within the rotation deviation distribution to obtain a fault deviation distribution, and a fault diagnosis result of the rotating equipment is identified and obtained.

[0008] In a second aspect, the present invention provides a rotating equipment fault diagnosis device, the device comprising: an equipment feature acquisition module, used to acquire equipment feature information of the rotating equipment to be fault diagnosed, wherein the equipment feature information includes weight distribution information and installation record information of the rotating equipment;

[0009] A basic deviation identification module is used to collect an operation image sequence and a vibration signal sequence of the rotating device during the rotation operation of the rotating device, perform rotation deviation identification according to the operation image sequence, and obtain a basic deviation distribution;

[0010] A deviation adjustment module, used to identify the rotation deviation according to the vibration signal sequence, adjust the basic deviation distribution, and obtain the rotation deviation distribution;

[0011] A fault diagnosis module is used to perform aging deviation analysis of the rotating equipment according to the equipment characteristic information to obtain aging deviation distribution, perform aging deviation extraction within the rotation deviation distribution to obtain fault deviation distribution, and identify and obtain fault diagnosis results of the rotating equipment.

[0012] In a third aspect, the present invention provides an electronic device, comprising: a memory for storing a computer software program; and a processor for reading and executing the computer software program, thereby implementing the rotating equipment fault diagnosis method of the first aspect.

[0013] In a fourth aspect, the present invention provides a non-transitory computer-readable storage medium having a computer software program stored therein, and when the computer software program is executed by a processor, the rotating equipment fault diagnosis method of the first aspect is implemented.

[0014] The present invention collects equipment characteristic information of the rotating equipment to be fault diagnosed, specifically including weight distribution information and installation record information of the rotating equipment, which can reflect the differentiated characteristics of the rotating equipment itself. During the rotating operation of the rotating equipment, the operating image sequence and vibration signal sequence of the rotating equipment are collected, and the rotation deviation is identified according to the operating image sequence to obtain the basic deviation distribution, and then the rotation deviation is identified according to the vibration signal sequence, and the basic deviation distribution is adjusted to obtain the rotation deviation distribution. The rotation deviation is identified from the two directions of image and signal characteristics to improve the accuracy of rotation deviation identification. According to the equipment characteristic information, the aging deviation analysis of the rotating equipment is performed to obtain the aging deviation distribution. The characteristics of the rotating equipment itself affect its aging during the working process. The aging deviation analysis is performed based on the equipment characteristic information, and the normal aging deviation caused by the operation of the equipment itself can be accurately obtained. Then, the aging deviation is extracted in the rotation deviation distribution to obtain the fault deviation distribution, and the fault diagnosis result of the rotating equipment is identified, and the influence of the specific aging deviation caused by the characteristics of the equipment itself is eliminated. The fault deviation diagnosis result caused by the fault is obtained, and accurate fault identification is completed, so as to achieve the technical effect of reflecting the actual deviation fault situation of the equipment and improving the accuracy of fault monitoring diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of a flow chart of a rotating equipment fault diagnosis method provided by the present invention;

[0016] Figure 2 A schematic diagram of the structure of the rotating equipment fault diagnosis device provided by the present invention;

[0017] Figure 3 A functional block diagram of the electronic device provided by the present invention;

[0018] Figure 4 A functional block diagram of a computer-readable storage device provided by the present invention.

[0019] In the accompanying drawings, the components represented by the reference numerals are listed as follows:

[0020] Equipment feature acquisition module 11, basic deviation identification module 12, deviation adjustment module 13, fault diagnosis module 14, electronic device 100, memory 110, processor 120, computer program 111, computer readable storage medium 200, computer program 211. DETAILED DESCRIPTION

[0021] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0022] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, "plurality" means two or more, unless otherwise clearly and specifically defined.

[0023] In the description of the present application, the term "for example" is used to mean "used as an example, illustration or description". Any embodiment described as "for example" in the present application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid unnecessary details to obscure the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present application.

[0024] Example 1

[0025] like Figure 1 As shown, an embodiment of the present invention provides a rotating equipment fault diagnosis method, comprising:

[0026] S10: collecting equipment characteristic information of the rotating equipment to be diagnosed for fault, wherein the equipment characteristic information includes weight distribution information and installation record information of the rotating equipment;

[0027] In the embodiments of the present application, the device characteristic information refers to data that can reflect the differentiated characteristics of the rotating device itself. There will be certain differences in the production, manufacturing and installation processes of each rotating device, which will have a certain impact on the deviation in the operation of the rotating device. For example, if the mass of the blades on one side of the fan is slightly greater than the mass of the other blades, the fan will deviate toward the side of the blade with slightly larger mass when it is running, causing the blade on the side with slightly larger mass to deviate away from the standard rotation center, and the opposite blade is close to the standard rotation center, forming an aging deviation. During the operation of the rotating device, other factors may also cause deviation from the rotation center, such as mechanical damage, external impact on the equipment, etc., forming a fault deviation.

[0028] For example, the rotating equipment is a rotating device such as a fan, a motor, etc. If the rotating equipment deviates from the rotation center, it will cause problems such as abnormal vibration of the equipment and fail to operate normally.

[0029] The equipment characteristic information of the rotating equipment to be diagnosed is collected. The equipment characteristic information includes the weight distribution information and installation record information of the rotating equipment. The uneven mass of each part of the rotating equipment in the weight distribution information will cause the rotation center to deviate. The installation record information includes information such as the installation position of the rotating equipment during installation, reflecting the slight deviation caused by inevitable errors during installation.

[0030] Step S10 in the method provided in the embodiment of the present application includes:

[0031] Collect weight information of multiple equipment areas of the rotating equipment to be fault diagnosed and construct weight distribution information;

[0032] Collecting installation images of the rotating device when it is installed;

[0033] An installation deviation analysis is performed on the installation image to obtain an installation deviation distribution as the installation record information, wherein a set of sample installation images is collected, and a sample installation deviation distribution set is obtained by annotating according to the deviation distances of the multiple device areas in each sample installation image from a preset rotation center, and an installation deviation analyzer is trained to perform installation deviation analysis.

[0034] In the embodiment of the present application, the weight distribution information refers to the weight data of each area of ​​the equipment, which is used to construct a mass distribution model of the entire rotating equipment.

[0035] Exemplarily, a precise mass measuring instrument (such as an electronic weighing system) is used to weigh multiple equipment areas of a rotating device one by one, and the weight of each area is recorded. Exemplarily, in the production process of a rotating device such as a fan, multiple equipment areas of the rotating device, such as multiple fan blades, are weighed separately to obtain weight information of the multiple equipment areas. Optionally, the density distribution of the multiple equipment areas can also be tested, and then the weight information of the multiple equipment areas is calculated in combination with the volume. The multiple equipment areas are divided according to the type of rotating equipment. Ideally, the weights of the multiple equipment areas are the same. If the rotating device is circular, the multiple equipment areas can be divided into multiple sectors.

[0036] Furthermore, the weight distribution information is constructed by identifying the positions of the multiple equipment areas and combining the weight information of the multiple equipment areas. For example, the weight distribution information includes the weight information of the first fan blade, the second fan blade, and so on.

[0037] In the embodiment of the present application, during the equipment installation process, a high-resolution camera is used to capture an installation image, and an image of the rotating equipment after installation is completed is recorded as the installation image. Optionally, by collecting images from multiple angles and multiple positions, it is ensured that the installation image fully covers the installation status of the equipment.

[0038] The installation deviation analysis process is performed on the installation image, specifically analyzing the deviation between the installation position of the rotating device and the preset rotation center, for example, analyzing the area of ​​each device and the preset rotation center of the installation, and then obtaining the deviation.

[0039] Exemplarily, based on multiple installation records of similar rotating equipment, sample installation images containing multiple equipment areas are collected to obtain a sample installation image set. And according to the distance of deviation of the multiple equipment areas from the preset rotation center in each sample installation image, manual or semi-automatic annotation is performed to obtain a sample installation deviation distribution set. For example, if the ideal distance between an equipment area and the preset rotation center (such as the outer edge of the fan blade) is 50 cm, and the measured distance is 49.5 cm, the deviation of -0.5 cm can be recorded in the sample installation deviation distribution, and the distance between other equipment areas and the preset rotation center can be tested and annotated to obtain a sample installation deviation distribution set.

[0040] Furthermore, a sample installation image set and a labeled sample installation deviation distribution set are used to train an installation deviation analyzer using a machine learning algorithm (such as a convolutional neural network (CNN)). Optionally, the sample installation image set is used as input data, the sample installation deviation distribution set is used as supervision data, a loss function such as a mean square error function is selected, and iterative training is performed to adjust the model parameters to minimize the loss function so that the output of the installation deviation analyzer is close to the supervision data until the installation deviation analyzer converges, for example, the accuracy meets a threshold, and the threshold is, for example, 92%, and the training is completed.

[0041] Based on the trained installation deviation analyzer, the current installation image is input, the image is analyzed and processed, and the output installation deviation distribution is obtained as the installation record information.

[0042] Based on the above steps, we can fully understand the installation status and quality distribution of the equipment, provide key data support for subsequent equipment operation monitoring and fault diagnosis, and accurately identify the deviation distribution caused by the differential characteristics of the equipment.

[0043] S20: During the rotation operation of the rotating device, collecting an operation image sequence and a vibration signal sequence of the rotating device, performing rotation deviation identification according to the operation image sequence, and obtaining a basic deviation distribution;

[0044] In the embodiment of the present application, during the rotational operation of the rotating equipment, the operation image sequence and vibration signal sequence of the equipment are collected in real time, and the rotation deviation is identified by analyzing the operation image sequence to obtain the basic deviation distribution of the rotating equipment reflected by the operation image.

[0045] The operation image and vibration signal of the rotating equipment can both reflect the deviation of the rotation center of the rotating equipment. For example, when the rotation center of the rotating equipment deviates, it will affect the rotation frequency signal of the rotating equipment. Based on the feature recognition of the rotation frequency signal, the deviation of the rotation center of the rotating equipment can be obtained.

[0046] Based on the operating image sequence, image recognition technology is used to identify the rotation center deviation reflected in the operating image and obtain the basic deviation distribution. Subsequently, the vibration signal sequence is combined for auxiliary identification to improve the accuracy of rotating equipment deviation identification.

[0047] In the embodiment of the present application, the running image sequence refers to the images continuously captured by the high-speed camera during the rotation of the device. For example, a high-speed camera can be installed around the rotating device to ensure that the camera's viewing angle can fully cover the rotating area of ​​the rotating device. When the device is running, the camera captures the rotation process of the device in real time at a high frame rate (such as 1000 frames per second) to generate a continuous image sequence and obtain a rotating image sequence.

[0048] The vibration signal sequence refers to the vibration data generated by the equipment during operation, which is collected by the vibration sensor. For practice, high-precision vibration sensors can be installed at key parts of the equipment (such as bearings and supporting structures). When the equipment is running, the sensor records the vibration signal in real time and transmits it to the computer that executes the method provided in the embodiment of this application through the data acquisition system. For example, a high-sensitivity vibration sensor (such as a MEMS sensor) is used in combination with a data acquisition system (such as an NI data acquisition card) to ensure the accuracy and real-time nature of the vibration data.

[0049] Step S20 in the method provided in the embodiment of the present application includes:

[0050] During the rotation of the rotating device, a rotation speed sequence of the rotating device is collected, wherein a plurality of rotation speeds in the rotation speed sequence correspond to timestamps of a plurality of operation images in the operation image sequence;

[0051] Performing rotation deviation identification according to the running image sequence to obtain multiple rotation deviation distributions;

[0052] According to the multiple rotation speeds, the multiple rotation deviation distributions are weightedly calculated to obtain a weighted rotation deviation distribution as a basic deviation distribution, wherein the size of the rotation speed is negatively correlated with the size of the weight.

[0053] In an embodiment of the present application, during the rotational operation of the rotating device, the rotation speed sequence and the operation image sequence of the device are synchronously acquired. The clarity of the operation images at different rotational speeds is different. By analyzing and weightedly calculating the rotational deviation distribution of the operation images at different rotational speeds, a more accurate basic deviation distribution is obtained.

[0054] The rotation speed sequence refers to the speed data of the device during the rotation process. The timestamp of these data corresponds to the image timestamp in the running image sequence. A high-precision rotation speed sensor can be installed on the shaft or other appropriate position of the rotating device to detect the rotation speed of the rotating device. When the rotating device is running, the sensor records the rotation speed of the device in real time, and attaches an accurate timestamp. Each rotation speed corresponds to the running image one by one. In other words, the multiple running images and multiple rotation speeds in the running image sequence and the rotation speed sequence are data collected simultaneously at multiple moments.

[0055] Rotation deviation identification is performed according to the operation image sequence to obtain multiple rotation deviation distributions, wherein the operation image can reflect the distance between multiple device areas in the rotating device and the preset rotation center, and then the rotation deviation distribution can be identified. In addition, the clarity of the operation images at different speeds is different, and the accuracy of the rotation deviation distribution obtained by identification is different. For example, the clarity of the operation image with a large rotation speed is lower. The rotation deviation identification is performed on the multiple operation images in the system to obtain multiple rotation deviation distributions.

[0056] Furthermore, according to the multiple rotation speeds, the multiple rotation deviation distributions are weightedly calculated to obtain a weighted rotation deviation distribution as a basic deviation distribution, wherein the size of the rotation speed is negatively correlated with the size of the weight. For example, the sum of the multiple rotation speeds is calculated, and then the weight is assigned according to the ratio of the inverse of the ratio of each rotation speed to the sum of the multiple rotation speeds to the sum of the multiple inverses, the size of the rotation speed is inversely proportional to the size of the weight, and the sum of the multiple weights is 1. Multiple rotation deviation distributions are weightedly calculated using multiple weights, for example, the distance between the same device area and the preset rotation center in the multiple rotation deviation distributions is weightedly calculated to obtain the weighted deviation distances of the multiple device areas to form a basic deviation distribution.

[0057] The step of "performing rotation deviation identification according to the running image sequence to obtain multiple rotation deviation distributions" in the method provided in the embodiment of the present application includes:

[0058] Collecting a sample running image set, dividing the sample running images at different running speeds, and obtaining a plurality of sample speed running image sets;

[0059] According to the deviation distances of multiple device areas of the rotating device in each sample speed running image from the preset rotation center, multiple sample speed deviation distribution sets are obtained by annotation;

[0060] Using the plurality of sample speed running image sets and the plurality of sample speed deviation distribution sets, training a plurality of rotation deviation recognition branches, and obtaining a rotation deviation identifier;

[0061] The multiple operating images are input into the rotation deviation identification branches corresponding to the multiple rotation speeds, and multiple rotation deviation distributions are obtained through analysis.

[0062] In the embodiment of the present application, by classifying and analyzing the operating images of the rotating equipment, an accurate rotation deviation identifier can be trained, and the operating images at different rotation speeds can be identified to obtain multiple rotation deviation distributions.

[0063] For example, during the operation of similar rotating equipment, a high-speed camera is used to capture a large number of operation images covering various operation speeds. According to the different operation speeds of the rotating equipment, the collected sample operation images are classified to form a plurality of sample speed operation image sets. For example, the classification can be performed according to different rotation speed intervals.

[0064] Furthermore, the preset rotation center position can be marked on the equipment design drawing, and for each sample speed operation image, the deviation distance between each equipment area and the preset rotation center is calculated, and the calculated deviation distance is marked on each image to form multiple sample speed deviation distribution sets. Each sample speed deviation distribution set refers to a deviation data set marked according to the deviation distance between each equipment area and the preset rotation center in the operation image. Multiple sample speed deviation distribution sets correspond one-to-one to multiple sample speed operation image sets.

[0065] Use multiple sample speeds to run image sets and multiple sample speed deviation distribution sets, and train multiple rotation deviation recognition branches. Among them, select a machine learning model suitable for image processing, such as a convolutional neural network (CNN), and train multiple rotation deviation recognition branches for different running speeds. Each branch independently identifies the rotation deviation at the corresponding speed. Sample data at different speeds are trained separately to improve training efficiency and convergence speed. Optionally, use a deep learning framework (such as TensorFlow, Keras, or PyTorch) to build and train multiple CNN models to ensure that each branch is optimized for data at different speeds. Through continuous training, the error of the rotation deviation recognition branch is reduced, and the recognition accuracy of the speed deviation distribution is improved.

[0066] After training multiple rotation deviation recognition branches, a rotation deviation identifier is obtained. For example, the input layers of multiple rotation deviation recognition branches are connected, and input rules corresponding to the rotation speed of the running image are set to input to the corresponding rotation deviation recognition branch to obtain the rotation deviation identifier.

[0067] The multiple running images actually collected are input into the rotation deviation recognition branches corresponding to the multiple rotation speeds. The trained CNN model is used to analyze and process the input image data, output accurate rotation deviation distribution data, and analyze to obtain multiple rotation deviation distributions.

[0068] Through the above steps, the operating images of rotating equipment at different speeds are accurately collected and classified, and by training multiple rotation deviation recognition branches, an accurate rotation deviation identifier is obtained. The rotation deviation distribution corresponding to multiple operating speeds is accurately analyzed and identified, providing important technical support and data basis for equipment fault diagnosis.

[0069] S30: performing rotation deviation identification according to the vibration signal sequence, adjusting the basic deviation distribution, and obtaining a rotation deviation distribution;

[0070] In an embodiment of the present application, multiple rotation deviation distributions obtained by rotation deviation identification based on the running image sequence may be inaccurate. Rotation deviation identification is further performed in combination with the vibration signal sequence to obtain a signal deviation distribution based on the vibration signal sequence analysis. The above-mentioned basic deviation distribution is adjusted to improve the accuracy of the deviation distribution analysis.

[0071] Among them, the vibration signal includes, for example, the rotation frequency signal of the rotating equipment, which is correlated with the amplitude of the deviation of the rotation center of the rotating equipment. The rotation deviation is identified based on the vibration signal sequence, and then based on the identified signal deviation distribution, a weighted calculation is performed in combination with the basic deviation distribution to adjust the basic deviation distribution to obtain the rotation deviation distribution combined with the comprehensive analysis of the image and signal.

[0072] For example, a machine learning method may be used to analyze the vibration signal to obtain a signal deviation distribution.

[0073] Step S30 in the method provided in the embodiment of the present application includes:

[0074] According to the deviation test data of the rotating equipment, a set of sample vibration signal sequences is collected and annotated to obtain a set of sample signal deviation distributions, wherein the sample vibration signal includes a rotation frequency signal;

[0075] Using the sample vibration signal sequence set and the sample signal deviation distribution set to train a signal deviation identifier;

[0076] Using the signal deviation identifier, performing rotation deviation identification on the vibration signal sequence to obtain a signal deviation distribution;

[0077] The signal deviation distribution is used to adjust the basic deviation distribution to obtain a rotation deviation distribution.

[0078] In an embodiment of the present application, when a similar rotating device is subjected to a deviation test, a high-precision vibration sensor is used to collect vibration signals in real time, and the device's rotation frequency signal is identified in the collected vibration signal, that is, a signal consistent with the rotation frequency of the rotating device, to form a set of sample vibration signal sequences.

[0079] According to the deviation of the actual rotation center of the rotating equipment test from the preset rotation center (i.e., the standard rotation center of the rotating equipment installation design), such as the deviation distance of multiple equipment areas from the preset rotation center, each sample vibration signal is labeled to obtain a sample signal deviation distribution set.

[0080] Optionally, during the signal acquisition process, the collected raw vibration data may be processed through vibration signal preprocessing to remove noise and interference and improve signal quality.

[0081] Furthermore, the sample vibration signal sequence set and the corresponding sample signal deviation distribution set are organized into a training data set. A machine learning model suitable for processing time series data, such as a long short-term memory network (LSTM) or a convolutional neural network (CNN), is selected to learn the implicit relationship of the data in the training data set. The signal deviation identifier is supervised and trained using the training data set, and the model parameters are adjusted according to the output error until it meets the convergence requirements, such as an accuracy rate of 90%, and the training converges to obtain a trained signal deviation identifier.

[0082] The signal deviation identifier is used to perform rotation deviation identification on the current vibration signal sequence, identify the deviation of the equipment, and obtain the deviation of multiple equipment areas as the signal deviation distribution.

[0083] Furthermore, the signal deviation distribution obtained based on the vibration signal sequence analysis is used to adjust the basic deviation distribution, so as to obtain the deviation of the rotating equipment by combining the vibration signal data and image data analysis.

[0084] Exemplarily, weighted calculation is performed on the deviation distances of multiple device areas in the signal deviation distribution and the basic deviation distribution from a preset rotation center to obtain weighted deviation distances of the multiple device areas, and to construct a rotation deviation distribution.

[0085] Among them, the weights of weighted calculation can be set based on technical personnel in this field, for example, based on the accuracy of basic deviation distribution based on running image analysis and the accuracy of signal deviation distribution based on vibration signal sequence analysis, weights are set and weighted calculation is performed.

[0086] The embodiment of the present application further identifies and adjusts the rotation deviation through the analysis of the vibration signal sequence and the training of the signal deviation identifier, and obtains a more accurate rotation deviation distribution. The advantages of image data and vibration data are combined to improve the accuracy and reliability of rotating equipment fault diagnosis.

[0087] S40: performing aging deviation analysis on the rotating equipment according to the equipment characteristic information to obtain aging deviation distribution, performing aging deviation extraction within the rotation deviation distribution to obtain fault deviation distribution, and identifying and obtaining fault diagnosis results of the rotating equipment.

[0088] In an embodiment of the present application, by performing aging deviation analysis on the equipment characteristic information, deviations caused by aging due to weight distribution and gravity during operation of the equipment can be identified, and by extracting the aging deviation within the rotation deviation distribution, the fault deviation distribution can be further obtained, and finally the fault diagnosis results of the rotating equipment can be identified.

[0089] Among them, the aging deviation caused by uneven weight and the aging deviation caused by gravity are analyzed according to the weight distribution information in the equipment characteristic information, and combined with the installation deviation distribution in the installation record information, as the aging deviation distribution, it reflects the reasonable deviation caused by factors such as aging during the installation and use of the rotating equipment. By extracting this part of the aging deviation distribution in the rotation deviation distribution, that is, deleting the aging deviation distribution, the fault deviation of the equipment caused by unreasonable use, external impact and other factors can be obtained, and then the fault deviation diagnosis can be performed.

[0090] Step S40 in the method provided in the embodiment of the present application includes:

[0091] Calculating and obtaining equipment weight information of the rotating equipment according to the weight distribution information;

[0092] Collecting the running time of the rotating equipment, and collecting the stop time set and the stop posture set of the rotating equipment, wherein the stop posture includes the equipment area closest to the ground when stopped;

[0093] performing weight distribution deviation identification according to the weight distribution information and the running time to obtain a weight distribution deviation distribution;

[0094] According to the equipment weight information, the outage time set and the outage posture set, equipment weight deviation is identified to obtain equipment weight deviation distribution;

[0095] The weight distribution deviation distribution, the equipment weight deviation distribution and the installation record information are integrated to obtain the aging deviation distribution.

[0096] In the embodiment of the present application, the weight data of each part is aggregated to calculate the total weight of the equipment, for example, the sum of the masses of multiple equipment areas is calculated to obtain the equipment weight information of the rotating equipment. Optionally, the weight of the rotating equipment after manufacturing is completed can also be tested to obtain the equipment weight information.

[0097] The running time of the rotating equipment is collected, and the stop time set and stop posture set of the rotating equipment are collected. The running time refers to the actual running time length of the rotating equipment, and the stop time set and stop posture set record the relevant information when the rotating equipment stops running.

[0098] For example, the equipment monitoring system records the cumulative running time of the equipment. Also, when the equipment is shut down, the shutdown time and shutdown posture are recorded. The shutdown posture includes the equipment area closest to the ground when shut down, that is, the equipment area at the bottom when shut down. When shut down, the rotating equipment will deviate slightly from the direction of gravity due to the influence of gravity. At this time, the equipment area at the bottom of the rotating equipment will deviate slightly from the preset rotation center. The shutdown time and shutdown posture will affect the deviation amplitude of the directions of different equipment areas due to the influence of gravity.

[0099] According to the weight distribution information and the operation time, the weight distribution deviation is identified to obtain the weight distribution deviation distribution, wherein the uneven weight of multiple equipment areas may cause the rotation center to deviate toward the heavier equipment area due to the difference in centrifugal force during the operation of the rotating equipment.

[0100] Exemplarily, a set of sample weight distribution information can be collected based on the rotation test data of similar rotating equipment, and a set of sample running time can be collected according to the test time. Then, the deviation changes caused by the rotating equipment's deviation from the initial installation distribution after different sample weight distribution information and sample running times are obtained as a set of sample weight distribution deviation distributions.

[0101] Optionally, based on machine learning, such as a feedforward neural network, a weight distribution deviation analyzer is constructed to predict the deviation change after running based on the weight distribution information and the running time, and then the sample weight distribution information set and the sample running time set are used as input data, and the sample weight distribution deviation distribution set is used as supervision data to train the weight distribution deviation analyzer. According to the deviation between the output and the supervision data, the parameters of the weight distribution deviation analyzer are adjusted. Each group of sample data undergoes multiple rounds of supervised training, and the model parameters are adjusted until the model converges, for example, the accuracy meets the requirements, and a trained weight distribution deviation analyzer is obtained.

[0102] Based on the trained weight distribution deviation analyzer, the current weight distribution information and running time are input and distributed and identified to obtain the weight distribution deviation distribution of the current rotating equipment due to the slight unevenness of the weight distribution information and the influence of the running time.

[0103] Furthermore, according to the equipment weight information, the outage time set and the outage posture set, the equipment weight deviation is identified to obtain the equipment weight deviation distribution. Among them, each time the equipment is out of service, the rotating equipment deviates uniformly toward the equipment area in the corresponding outage posture. Combining the outage time and equipment weight information, the deviation of the rotating equipment caused by the outage can be obtained.

[0104] Exemplarily, based on the equipment weight information, downtime information, etc. in the test data of similar rotating equipment, a set of sample equipment weight information and a set of sample downtime time can be collected, and the distance that the rotating equipment deviates in the direction of gravity under different sample equipment weight information and sample downtime can be collected to obtain a set of sample equipment weight deviations.

[0105] Optionally, a classifier of equipment weight deviation can be constructed by using a set of sample equipment weight information, a set of sample outage time, and a set of sample equipment weight deviation. The current equipment weight information and outage time are input into the classifier, and the corresponding equipment weight deviation distance can be obtained by classification. In combination with the equipment area directly below the corresponding outage posture, the corresponding equipment weight deviation distribution can be obtained. Optionally, the unit equipment weight deviation distance of the unit equipment weight under the unit time can be calculated according to the set of sample equipment weight information, the set of sample outage time, and the set of sample equipment weight deviation, so as to calculate the equipment weight deviation of the current equipment weight information and outage time.

[0106] Based on the outage time set and outage posture set of the current multiple outages, combined with the equipment weight information, the equipment weight deviation of the multiple outages can be calculated, and then the total deviation distance of multiple equipment areas can be calculated and added to obtain the equipment weight deviation distribution. For example, an equipment area is located closest to the ground and farthest from the ground in two outages, and the corresponding equipment deviation distances are -0.5mm (close to the preset rotation center) and 0.4mm (far away from the preset rotation center), respectively, then the total deviation distance of the equipment area is -0.1mm.

[0107] Further, the weight distribution deviation distribution, the equipment weight deviation distribution and the installation deviation distribution in the installation record information are integrated to obtain the aging deviation distribution. Exemplarily, the sum of the deviation distances of each equipment area from the rotation center in the fused weight distribution deviation distribution, the equipment weight deviation distribution and the installation deviation distribution is calculated to obtain the total deviation distances of multiple equipment areas as the aging deviation distribution. Exemplarily, if the deviation distances of a device area from the preset rotation center in the weight distribution deviation distribution, the equipment weight deviation distribution and the installation deviation distribution are -0.3mm, -0.1mm and 0.2mm, then the total aging deviation distance of the device area is calculated to be -0.2mm.

[0108] Through the above steps, the deviation distribution of rotating equipment caused by uneven weight, gravity and slight deviation during installation is comprehensively analyzed during the total use process, and various information is integrated to form the aging deviation distribution. Combining various data analysis and machine learning technologies, this part of normal aging deviation can be extracted later to perform fault deviation diagnosis, which improves the accuracy and reliability of equipment aging analysis and fault diagnosis, and provides solid technical support for equipment maintenance and management.

[0109] Step S40 in the method provided in the embodiment of the present application further includes:

[0110] Extracting and deleting aging deviations of multiple equipment regions in the aging deviation distribution in the rotation deviation distribution to obtain a fault deviation distribution;

[0111] According to the fault deviation distribution, deviation fault identification is performed to obtain a deviation fault level;

[0112] According to the proportion of the aging deviation distribution in the fault deviation distribution, a fault correction coefficient is set, the deviation fault level is corrected, and a corrected fault level is obtained as a fault diagnosis result.

[0113] In the embodiment of the present application, by extracting and deleting the aging deviation distribution within the rotation deviation distribution, it is possible to obtain the fault deviation distribution caused by external impact and improper use, and further perform deviation fault identification and level correction, ultimately obtaining an accurate fault diagnosis result.

[0114] Among them, the fault deviation distribution refers to identifying the real fault deviation of the equipment by extracting and deleting the aging deviation distribution in the rotation deviation distribution. Deviation fault identification is based on the fault deviation distribution to identify and evaluate the fault condition of the equipment and obtain the deviation fault level. Deviation fault identification can be evaluated based on the rotating equipment fault evaluation standard in the prior art. For example, the greater the deviation distance, the greater the fault level.

[0115] The rotation deviation distribution includes rotation deviation distances of multiple equipment areas, which include aging deviation and fault deviation. The aging deviations of multiple equipment areas in the aging deviation distribution are extracted and deleted to obtain the fault deviations of the multiple equipment areas.

[0116] For example, if the rotation deviation distance between a device area and the preset rotation center is -1mm, and the aging deviation distance is -0.3mm, then the fault deviation distance between the device area and the preset rotation center is -1-(-0.3) = -0.7mm. In this way, the fault deviation distances of multiple device areas can be extracted and calculated to obtain the fault deviation distribution.

[0117] According to the fault deviation distribution, deviation fault identification is performed to obtain the deviation fault level. For example, a comparison table of different fault deviation distance intervals and fault levels can be constructed based on the deviation fault data record log of the rotating equipment. The fault level is, for example, 5 levels from 1 to 5. The greater the deviation distance, the greater the deviation fault level. The maximum fault deviation distance in the current fault deviation distribution is input into the comparison table to obtain the deviation fault level.

[0118] Optionally, methods such as machine learning may be selected to deviate from fault data.

[0119] Furthermore, according to the proportion of the aging deviation distribution in the fault deviation distribution, a fault correction coefficient is set to correct the deviation fault level. The greater the proportion of the aging deviation distribution in the fault deviation distribution, the smaller the fault deviation amplitude of the rotating equipment and the smaller the fault deviation degree of the rotating equipment. Conversely, the greater the fault deviation amplitude of the rotating equipment and the greater the fault deviation degree of the rotating equipment.

[0120] Exemplarily, according to the proportion of the aging deviation distribution in the fault deviation distribution, for example, a ratio of the maximum aging deviation distance in the aging deviation distribution to the maximum fault deviation distance in the fault deviation distribution is calculated as the proportion.

[0121] The larger the proportion, the smaller the fault correction coefficient is set to be, and the deviation fault level is multiplied and corrected to reduce the deviation fault level with a smaller fault deviation proportion. Conversely, the larger the fault correction coefficient is set to be, the larger the fault correction coefficient is set to be, and the deviation fault level is multiplied and corrected to increase the deviation fault level with a larger fault deviation proportion, so as to facilitate the correction of the inspection and maintenance attention of different fault deviation degrees.

[0122] Optionally, based on the steps in the aforementioned content, multiple proportions can be determined based on the deviation monitoring data of multiple similar rotating equipment, and the average proportion can be calculated. The inverse of the ratio of the current proportion to the average proportion is used as the fault correction coefficient, and the deviation fault level is multiplied and rounded to complete the correction and obtain the corrected fault level as the fault diagnosis result.

[0123] Through the above steps, the embodiment of the present application accurately identifies and evaluates the deviation fault of the rotating equipment, considers the impact of aging deviation on fault deviation, performs extraction and fault analysis, and corrects and generates a corrected fault level, thereby improving the accuracy and reliability of fault diagnosis.

[0124] The embodiments of the present application achieve the following technical effects through the above specific implementation methods:

[0125] The present invention collects device characteristic information of the rotating equipment to be fault diagnosed, specifically including weight distribution information and installation record information of the rotating equipment, which can reflect the differentiated characteristics of the rotating equipment itself. During the rotating operation of the rotating equipment, the operating image sequence and vibration signal sequence of the rotating equipment are collected, and the rotation deviation is identified according to the operating image sequence to obtain the basic deviation distribution, and then the rotation deviation is identified according to the vibration signal sequence, and the basic deviation distribution is adjusted to obtain the rotation deviation distribution. The rotation deviation is identified from the two directions of image and signal characteristics to improve the accuracy of rotation deviation identification. According to the device characteristic information, the aging deviation analysis of the rotating equipment is performed to obtain the aging deviation distribution. The characteristics of the rotating equipment itself affect its aging during the working process. The aging deviation analysis is performed based on the device characteristic information, and the normal aging deviation caused by the operation of the device itself can be accurately obtained. Then, the aging deviation is extracted in the rotation deviation distribution to obtain the fault deviation distribution, and the fault diagnosis result of the rotating equipment is identified, and the influence of the specific aging deviation caused by the characteristics of the equipment itself is eliminated. The fault deviation diagnosis result caused by the fault is obtained, and accurate fault identification is completed, so as to achieve the technical effect of reflecting the actual deviation fault situation of the equipment and improving the accuracy of fault monitoring diagnosis.

[0126] Example 2

[0127] like Figure 2 As shown, based on the same inventive concept as the rotating equipment fault diagnosis method provided in Example 1, an embodiment of the present invention provides a rotating equipment fault diagnosis device, the device comprising:

[0128] The equipment characteristic acquisition module 11 is used to acquire equipment characteristic information of the rotating equipment to be diagnosed for fault, wherein the equipment characteristic information includes weight distribution information and installation record information of the rotating equipment;

[0129] A basic deviation identification module 12 is used to collect an operation image sequence and a vibration signal sequence of the rotating device during the rotation operation of the rotating device, perform rotation deviation identification according to the operation image sequence, and obtain a basic deviation distribution;

[0130] A deviation adjustment module 13, configured to identify rotation deviation according to the vibration signal sequence, adjust the basic deviation distribution, and obtain a rotation deviation distribution;

[0131] The fault diagnosis module 14 is used to perform aging deviation analysis of the rotating equipment according to the equipment characteristic information, obtain aging deviation distribution, extract aging deviation within the rotation deviation distribution, obtain fault deviation distribution, and identify and obtain the fault diagnosis result of the rotating equipment.

[0132] Furthermore, the device feature collection module 11 is also used for:

[0133] Collect weight information of multiple equipment areas of the rotating equipment to be fault diagnosed and construct weight distribution information;

[0134] Collecting installation images of the rotating device when it is installed;

[0135] An installation deviation analysis is performed on the installation image to obtain an installation deviation distribution as the installation record information, wherein a set of sample installation images is collected, and a sample installation deviation distribution set is obtained by annotating according to the deviation distances of the multiple device areas in each sample installation image from a preset rotation center, and an installation deviation analyzer is trained to perform installation deviation analysis.

[0136] Furthermore, the basic deviation identification module 12 is also used for:

[0137] During the rotation of the rotating device, a rotation speed sequence of the rotating device is collected, wherein a plurality of rotation speeds in the rotation speed sequence correspond to timestamps of a plurality of operation images in the operation image sequence;

[0138] Performing rotation deviation identification according to the running image sequence to obtain multiple rotation deviation distributions;

[0139] According to the multiple rotation speeds, the multiple rotation deviation distributions are weightedly calculated to obtain a weighted rotation deviation distribution as a basic deviation distribution, wherein the size of the rotation speed is negatively correlated with the size of the weight.

[0140] The rotation deviation is identified according to the running image sequence to obtain multiple rotation deviation distributions, including:

[0141] Collecting a sample running image set, dividing the sample running images at different running speeds, and obtaining a plurality of sample speed running image sets;

[0142] According to the deviation distances of multiple device areas of the rotating device in each sample speed running image from the preset rotation center, multiple sample speed deviation distribution sets are obtained by annotation;

[0143] Using the plurality of sample speed running image sets and the plurality of sample speed deviation distribution sets, training a plurality of rotation deviation recognition branches, and obtaining a rotation deviation identifier;

[0144] The multiple operating images are input into the rotation deviation identification branches corresponding to the multiple rotation speeds, and multiple rotation deviation distributions are obtained through analysis.

[0145] Furthermore, the deviation adjustment module 13 is also used for:

[0146] According to the deviation test data of the rotating equipment, a set of sample vibration signal sequences is collected and annotated to obtain a set of sample signal deviation distributions, wherein the sample vibration signal includes a rotation frequency signal;

[0147] Using the sample vibration signal sequence set and the sample signal deviation distribution set to train a signal deviation identifier;

[0148] Using the signal deviation identifier, performing rotation deviation identification on the vibration signal sequence to obtain a signal deviation distribution;

[0149] The signal deviation distribution is used to adjust the basic deviation distribution to obtain a rotation deviation distribution.

[0150] Furthermore, the fault diagnosis module 14 is also used for:

[0151] Calculating and obtaining equipment weight information of the rotating equipment according to the weight distribution information;

[0152] Collecting the running time of the rotating equipment, and collecting the stop time set and the stop posture set of the rotating equipment, wherein the stop posture includes the equipment area closest to the ground when stopped;

[0153] performing weight distribution deviation identification according to the weight distribution information and the running time to obtain a weight distribution deviation distribution;

[0154] According to the equipment weight information, the outage time set and the outage posture set, equipment weight deviation is identified to obtain equipment weight deviation distribution;

[0155] The weight distribution deviation distribution, the equipment weight deviation distribution and the installation record information are integrated to obtain the aging deviation distribution.

[0156] Furthermore, the fault diagnosis module 14 is also used for:

[0157] Extracting and deleting aging deviations of multiple equipment regions in the aging deviation distribution in the rotation deviation distribution to obtain a fault deviation distribution;

[0158] According to the fault deviation distribution, deviation fault identification is performed to obtain a deviation fault level;

[0159] According to the proportion of the aging deviation distribution in the fault deviation distribution, a fault correction coefficient is set, the deviation fault level is corrected, and a corrected fault level is obtained as a fault diagnosis result.

[0160] See also Figure 3 , Figure 3 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 3As shown, an embodiment of the present invention provides an electronic device 100, including a memory 110, a processor 120, and a computer program 111 stored in the memory 110 and executable on the processor 120. When the processor 120 executes the computer program 111, the steps of the method in Example 1 are implemented.

[0161] See also Figure 4 , Figure 4 A schematic diagram of an embodiment of a computer-readable storage medium provided in an embodiment of the present invention. Figure 4 As shown, this embodiment provides a computer-readable storage medium 200 on which a computer program 211 is stored. When the computer program 211 is executed by a processor, the steps of the method in Embodiment 1 are implemented.

[0162] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0163] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0164] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0165] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0166] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0167] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0168] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for diagnosing faults of rotating equipment, characterized in that: The method comprises: Collecting equipment characteristic information of the rotating equipment to be diagnosed for fault, wherein the equipment characteristic information includes weight distribution information and installation record information of the rotating equipment; During the rotation operation of the rotating device, an operation image sequence and a vibration signal sequence of the rotating device are collected, and rotation deviation is identified according to the operation image sequence to obtain a basic deviation distribution; Performing rotation deviation identification according to the vibration signal sequence, adjusting the basic deviation distribution, and obtaining a rotation deviation distribution; According to the equipment characteristic information, an aging deviation analysis of the rotating equipment is performed to obtain an aging deviation distribution, an aging deviation extraction is performed within the rotation deviation distribution to obtain a fault deviation distribution, and a fault diagnosis result of the rotating equipment is identified and obtained.

2. The rotating equipment fault diagnosis method according to claim 1, characterized in that: Collect equipment characteristic information of the rotating equipment to be diagnosed, including: Collect weight information of multiple equipment areas of the rotating equipment to be fault diagnosed and construct weight distribution information; Collecting installation images of the rotating device when it is installed; An installation deviation analysis is performed on the installation image to obtain an installation deviation distribution as the installation record information, wherein a set of sample installation images is collected, and a sample installation deviation distribution set is obtained by annotating according to the deviation distances of the multiple device areas in each sample installation image from a preset rotation center, and an installation deviation analyzer is trained to perform installation deviation analysis.

3. The rotating equipment fault diagnosis method according to claim 1, characterized in that: Performing rotation deviation identification according to the running image sequence to obtain basic deviation distribution includes: During the rotation of the rotating device, a rotation speed sequence of the rotating device is collected, wherein a plurality of rotation speeds in the rotation speed sequence correspond to timestamps of a plurality of operation images in the operation image sequence; Performing rotation deviation identification according to the running image sequence to obtain multiple rotation deviation distributions; According to the multiple rotation speeds, the multiple rotation deviation distributions are weightedly calculated to obtain a weighted rotation deviation distribution as a basic deviation distribution, wherein the size of the rotation speed is negatively correlated with the size of the weight.

4. The rotating equipment fault diagnosis method according to claim 3, characterized in that: Rotation deviation identification is performed according to the running image sequence to obtain multiple rotation deviation distributions, including: Collecting a sample running image set, dividing the sample running images at different running speeds, and obtaining a plurality of sample speed running image sets; According to the deviation distances of multiple device areas of the rotating device in each sample speed running image from the preset rotation center, multiple sample speed deviation distribution sets are obtained by annotation; Using the plurality of sample speed running image sets and the plurality of sample speed deviation distribution sets, training a plurality of rotation deviation recognition branches, and obtaining a rotation deviation identifier; The multiple operating images are input into the rotation deviation identification branches corresponding to the multiple rotation speeds, and multiple rotation deviation distributions are obtained through analysis.

5. The rotating equipment fault diagnosis method according to claim 1, characterized in that: Performing rotation deviation identification according to the vibration signal sequence and adjusting the basic deviation distribution to obtain the rotation deviation distribution includes: According to the deviation test data of the rotating equipment, a set of sample vibration signal sequences is collected and annotated to obtain a set of sample signal deviation distributions, wherein the sample vibration signal includes a rotation frequency signal; Using the sample vibration signal sequence set and the sample signal deviation distribution set to train a signal deviation identifier; Using the signal deviation identifier, performing rotation deviation identification on the vibration signal sequence to obtain a signal deviation distribution; The signal deviation distribution is used to adjust the basic deviation distribution to obtain a rotation deviation distribution.

6. The rotating equipment fault diagnosis method according to claim 1, characterized in that: Performing aging deviation analysis of the rotating equipment according to the equipment characteristic information to obtain aging deviation distribution includes: Calculating and obtaining equipment weight information of the rotating equipment according to the weight distribution information; Collecting the running time of the rotating equipment, and collecting the stop time set and the stop posture set of the rotating equipment, wherein the stop posture includes the equipment area closest to the ground when stopped; performing weight distribution deviation identification according to the weight distribution information and the running time to obtain a weight distribution deviation distribution; According to the equipment weight information, the outage time set and the outage posture set, equipment weight deviation is identified to obtain equipment weight deviation distribution; The weight distribution deviation distribution, the equipment weight deviation distribution and the installation record information are integrated to obtain the aging deviation distribution.

7. The rotating equipment fault diagnosis method according to claim 1, characterized in that: Extracting aging deviations from the rotation deviation distribution to obtain fault deviation distribution, and identifying fault diagnosis results of the rotating equipment, including: Extracting and deleting aging deviations of multiple equipment regions in the aging deviation distribution in the rotation deviation distribution to obtain a fault deviation distribution; According to the fault deviation distribution, deviation fault identification is performed to obtain a deviation fault level; According to the proportion of the aging deviation distribution in the fault deviation distribution, a fault correction coefficient is set, the deviation fault level is corrected, and a corrected fault level is obtained as a fault diagnosis result.

8. A rotating equipment fault diagnosis device, characterized in that: The device comprises: An equipment feature collection module, used to collect equipment feature information of the rotating equipment to be diagnosed for fault, wherein the equipment feature information includes weight distribution information and installation record information of the rotating equipment; A basic deviation identification module is used to collect an operation image sequence and a vibration signal sequence of the rotating device during the rotation operation of the rotating device, perform rotation deviation identification according to the operation image sequence, and obtain a basic deviation distribution; A deviation adjustment module, used to identify the rotation deviation according to the vibration signal sequence, adjust the basic deviation distribution, and obtain the rotation deviation distribution; A fault diagnosis module is used to perform aging deviation analysis of the rotating equipment according to the equipment characteristic information to obtain aging deviation distribution, perform aging deviation extraction within the rotation deviation distribution to obtain fault deviation distribution, and identify and obtain fault diagnosis results of the rotating equipment.

9. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing the rotating equipment fault diagnosis method described in any one of claims 1-7.

10. A non-transitory computer-readable storage medium, characterized in that: The storage medium stores a computer software program, and when the computer software program is executed by the processor, the rotating equipment fault diagnosis method according to any one of claims 1 to 7 is implemented.