CN111220373A - Method for diagnosing faults of centrifugal pump rotor system - Google Patents
Method for diagnosing faults of centrifugal pump rotor system Download PDFInfo
- Publication number
- CN111220373A CN111220373A CN202010192412.5A CN202010192412A CN111220373A CN 111220373 A CN111220373 A CN 111220373A CN 202010192412 A CN202010192412 A CN 202010192412A CN 111220373 A CN111220373 A CN 111220373A
- Authority
- CN
- China
- Prior art keywords
- fault
- mnj
- centrifugal pump
- pump rotor
- rotor system
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
- G01M13/04—Bearings
- G01M13/045—Acoustic or vibration analysis
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Monitoring And Testing Of Nuclear Reactors (AREA)
Abstract
The invention discloses a method for diagnosing faults of a centrifugal pump rotor system, particularly relates to a method for diagnosing faults of a centrifugal pump rotor system based on variational modal decomposition and a nuclear limit learning machine, and belongs to the field of fault diagnosis of rotary machinery. The method mainly comprises the following steps of S1: and acquiring vibration acceleration signals (normal, rotor misalignment, rotor unbalance, bearing inner ring fault, bearing outer ring fault and bearing rolling body fault) of the centrifugal pump rotor system in a normal state and a fault state to obtain a time domain signal sample set. Step S2: and carrying out variation modal decomposition on the obtained time domain signal sample set to obtain an eigenmode function component. Step S3: energy values, form factors, pulse indexes, margin coefficients, peak factors and kurtosis values are obtained for each eigenmode function component. Step S4: and (5) constructing a feature matrix and normalizing the processed data. Step S5: and training the fault diagnosis system of the centrifugal pump rotor system by adopting the training sample. Step S6: and inputting the test sample or the real-time sample into a fault diagnosis model of the centrifugal pump rotor system to diagnose the fault.
Description
Technical Field
The invention belongs to the field of fault diagnosis of rotating machinery, and particularly relates to a fault diagnosis method for a centrifugal pump rotor system based on variational modal decomposition and a nuclear limit learning machine.
Background
The centrifugal pump is a general rotating mechanical product and is widely applied to various industries including petrochemical industry, water supply and drainage, agricultural irrigation and the like. Statistically, about 21% of the electricity is applied to the driving of various types of pumps, and the centrifugal pump is the most used one of the pumps. The centrifugal pump rotor system mainly comprises an impeller, a rolling bearing, a shaft, a coupling and other parts, and the rotor is the part with the highest failure rate of the centrifugal pump. Therefore, the condition monitoring and fault diagnosis are carried out on the centrifugal pump rotor system, so that the safe and reliable operation of the centrifugal pump rotor system is ensured, and the method has important significance for improving the safety performance of production.
At present, the most common method for monitoring and diagnosing the state of the centrifugal pump is to monitor meters such as flow pressure of a centrifugal pump unit to judge faults, but the faults need to acquire random vibration signals of the centrifugal pump when being accurately and efficiently diagnosed, and then the random vibration signals are subjected to time-frequency analysis. Due to the characteristics of complex fault information components, micro fault characteristic information, low signal-to-noise ratio and the like of the actually measured vibration signal, efficient and accurate judgment is difficult.
Empirical Mode Decomposition (EMD) is a new method suitable for processing nonlinear, non-stationary signals, and essentially consists of smoothing the signal and decomposing a complex signal into a series of natural modal components. Compared with other analysis methods, the empirical mode decomposition is a self-adaptive time-frequency analysis method without any prior knowledge, has self-adaptive signal decomposition and noise reduction capabilities, and is successfully applied to the field of fault diagnosis of rotating machinery such as centrifugal pumps, fans and the like at present. However, the decomposition process of empirical mode decomposition has problems of unstable decomposition, end point effect and the like. To solve this problem, a Local Mean Decomposition (LMD) method has been proposed. He adaptively decomposes the original signal into a form of summation of a plurality of function products, which not only can obtain the video distribution of the signal, but also can obtain the instantaneous frequency with physical significance. However, the problems of end-point effect and mode aliasing still cannot be thoroughly solved in the decomposition of noisy signals and early-stage failure signals. In order to solve the problem, variable mode decomposition (VMD for short) is proposed, a frequency domain iteration mode is adopted to search an optimal solution of a variable model to determine the center frequency and the bandwidth of each component, the components are separated through self-adaptive subdivision of a frequency domain, the method is applied to mechanical fault diagnosis, and the variable mode decomposition method can effectively separate harmonic signals with close frequencies and frequency mutation signals and is suitable for separation of multi-component non-stationary signals. The method has the advantages over the previous adaptive decomposition method in that the problem of continuous accumulation of envelope estimation errors is solved, and the endpoint effect can be effectively overcome.
At present, a variation modal mode decomposition method is used, most of the method is to decompose a signal, screen out a valuable eigenmode function and reconstruct the signal. A method for extracting information from each eigenmode function is provided, and each proposed information is combined into a fault feature for further fault diagnosis.
Extreme learning machines are machine learning systems or methods constructed based on feedforward neural networks, are suitable for supervised learning and unsupervised learning problems, and have been successfully applied to computer vision, biological information and the like. The traditional feedforward neural network adopts a gradient descent iterative algorithm to challenge the weight parameters, which has obvious defects: the learning speed is slow, so that the system computing time is increased; the learning rate is difficult to determine and tends to fall into a local minimum; it is easy to bleed and over-train, causing the generalization performance to be reduced. These deficiencies become application bottlenecks that limit the use of feedforward neural networks using iterative algorithms. In response to these problems, extreme learning algorithms have evolved. The problem of the classification limitation of the small samples of the support vector machine can be solved. By introducing the sum function, the extreme learning machine algorithm can be improved into a kernel extreme learning machine, hidden layer parameters do not need to be adjusted repeatedly, the traditional single hidden layer feedforward neural network parameter training problem is converted into a linear equation solving problem, the minimum norm least square solution obtained is used as a network output weight, and the whole training process is completed at one time. Therefore, the training speed is greatly improved, and the generalization performance is better.
Disclosure of Invention
The invention provides a centrifugal pump rotor system fault diagnosis method based on variational modal decomposition and a nuclear limit learning machine, aiming at the problems that the centrifugal pump rotor system has multiple fault types, and fault signals have non-stable and non-linear characteristics and are difficult to identify.
The invention adopts the technical scheme that the fault diagnosis method is a fault diagnosis method of variational modal decomposition and mode identification of a nuclear extreme learning machine, and the fault diagnosis method of the centrifugal pump rotor system comprises the following steps:
step S1: respectively acquiring n groups of vibration acceleration signals, n, of a centrifugal pump rotor system in 6 states of normal state, unbalanced rotor state, rotor misalignment state, bearing inner ring fault state, bearing outer ring fault state and bearing rolling element fault state>The six states respectively obtain n groups of time domain signal sample sets, which are respectively marked as xmn(t), wherein m is 1, 2, 3, 4, 5, 6; n is 1, 2, 3, n.
Step S2: and (3) carrying out Variational Modal Decomposition (VMD) on each group of vibration time domain signal sample sets in six states to respectively obtain i eigenmode functions, namely IMF components. Denote each IMF component as cmni(t) then
Wherein i is IMFmnNumber of component, j being IMFmnThe number of components.
Step S3: for each IMFmnJ of a component cmni(t) calculating energy value, form factor, pulse index, margin coefficient, peak factor and kurtosis value, respectively recording as Emnj、Smnj、Imnj、Lmnj、Cmnj、Kmnj。
Wherein, XrmsIs the IMF component root mean square;
is the IMF component average amplitude; xpIs the IMF componentA peak value;
is the IMF component mean; xrthe square root amplitude of the IMF component, and β the kurtosis of the IMF component.
Step S4: and constructing a feature matrix.
Step S4.1: energy value EmnjWave form factor SmnjPulse index ImnjMargin coefficient LmnjCrest factor CmnjKurtosis value KmnjA fault signature matrix Z of (mn) xj is formed,
Z＝[ZE，ZS，ZI，ZL，ZC，ZK，]
step S4.2: and (5) normalizing the feature matrix Z data. The max-min normalization is used by linearly transforming the matrix data to map the original values to the interval [0,1] by max-min normalization.
Wherein x' is the new data after normalization; x is the raw data; minA is the minimum value in the data set; max a is the maximum value in the data set.
Step S4.3: and dividing the feature matrix after the normalization processing into a training sample and a test sample.
Step S5: and training the fault diagnosis system of the centrifugal pump rotor system by adopting the training sample.
Step S5.1: the extreme learning machine model is trained using training samples.
Step S5.2: and optimizing the extreme learning machine model by adopting a kernel function, wherein the optimized training model becomes a kernel extreme learning machine.
Step S6: and inputting the test sample or the real-time sample into a fault diagnosis model of the centrifugal pump rotor system to diagnose the fault.
The invention provides a method for diagnosing faults of a centrifugal pump rotor system based on variational modal decomposition and a nuclear extreme learning machine, which adopts variational modal decomposition to decompose an original vibration signal of a centrifugal pump into a group of eigenmode function components, calculates an energy value, a wave form factor, a pulse index, a margin coefficient, a peak value factor and a kurtosis value of each eigenmode function component, forms a characteristic vector, and is used as a training set and a testing set of the extreme learning machine after normalization processing, thereby improving the accuracy of fault diagnosis. The invention adopts the kernel function optimization extreme learning machine model, and the kernel extreme learning machine model has higher fault diagnosis rate, shortens the diagnosis time and improves the generalization of the model.
Drawings
FIG. 1 is a flowchart of the overall steps of the present invention.
Detailed Description
To further illustrate the technical solutions adopted to achieve the intended purposes of the present invention, please refer to the following detailed description of the present invention and the accompanying drawings, which are provided for reference and illustration only and are not intended to limit the present invention. As shown in fig. 1, the method for diagnosing a fault of a centrifugal pump rotor system based on variational modal decomposition and a nuclear limit learning machine includes obtaining a vibration acceleration signal, performing variational modal decomposition, calculating eigenmode function component calculation, constructing a fault feature matrix, establishing a fault diagnosis model of the centrifugal pump rotor system based on the nuclear limit learning machine, and performing fault diagnosis. The method comprises the following specific steps:
step S1: respectively acquiring n groups of vibration acceleration signals, n, of a centrifugal pump rotor system in 6 states of normal state, unbalanced rotor state, rotor misalignment state, bearing inner ring fault state, bearing outer ring fault state and bearing rolling element fault state>The six states respectively obtain n groups of time domain signal sample sets, which are respectively marked as xmn(t), wherein m is 1, 2, 3, 4, 5, 6; n is 1, 2, 3, n.
Step S2: and (3) carrying out Variational Modal Decomposition (VMD) on each group of vibration time domain signal sample sets in six states to respectively obtain i eigenmode functions, namely IMF components. Denote each IMF component as cmni(t) then
Wherein i is IMFmnNumber of component, j being IMFmnThe number of components.
Step S3: for each IMFmnJ of a component cmni(t) calculating energy value, form factor, pulse index, margin coefficient, peak factor and kurtosis value, respectively recording as Emnj、Smnj、Imnj、Lmnj、Cmnj、Kmnj。
Wherein, XrmsIs the IMF component root mean square;
is the IMF component average amplitude; xpIs the IMF component peak;
is the IMF component mean; xrthe square root amplitude of the IMF component, and β the kurtosis of the IMF component.
Step S4: constructing feature matrices
Step S4.1: energy value EmnjWave form factor SmnjPulse index ImnjMargin coefficient LmnjCrest factor CmnjKurtosis value KmnjA fault signature matrix Z of (mn) xj is formed,
Z＝[ZE，ZS，ZI，ZL，ZC，ZK，]
step S4.2: and (5) normalizing the feature matrix Z data. The max-min normalization is used by linearly transforming the matrix data to map the original values to the interval [0,1] by max-min normalization.
Wherein x' is the new data after normalization; x is the raw data; minA is the minimum value in the data set; max a is the maximum value in the data set.
Step S4.3: and dividing the feature matrix after the normalization processing into a training sample and a test sample.
Step S5: and (3) carrying out novel training on the fault diagnosis system of the centrifugal pump rotor system by adopting a training sample.
Step S5.1: the extreme learning machine model is trained using training samples.
Step S5.2: and optimizing the extreme learning machine model by adopting a kernel function, wherein the optimized training model becomes a kernel extreme learning machine.
Step S6: and inputting the test sample or the real-time sample into a fault diagnosis model of the centrifugal pump rotor system to diagnose the fault.
Claims (3)
1. A method for diagnosing faults of a centrifugal pump rotor system is characterized by comprising the following steps: the method comprises the following steps of,
step S1; respectively acquiring n groups of vibration acceleration signals, n, of a centrifugal pump rotor system in six states of normal state, unbalanced rotor state, misaligned rotor state, bearing inner ring fault state, bearing outer ring fault state and bearing rolling element fault state>The six states respectively obtain n groups of time domain signal sample sets, which are respectively marked as xmn(t), wherein m is 1, 2, 3, 4, 5, 6; n-1, 2, 3, n;
step S2: performing Variational Modal Decomposition (VMD) on each group of vibration time domain signal sample sets in six states to respectively obtain i eigenmode functions, namely IMF components;
step S3: for each IMFmnJ of a component cmni(t) calculating energy value, form factor, pulse index, margin coefficient, peak factor and kurtosis value, respectively recording as Emnj、Smnj、Imnj、Lmnj、Cmnj、Kmnj；
Step S4: constructing a feature matrix;
step S5: training a fault diagnosis system of a centrifugal pump rotor system by using a training sample;
step S6: and inputting the test sample or the real-time sample into a fault diagnosis model of the centrifugal pump rotor system to diagnose the fault.
2. The method for diagnosing the fault of the centrifugal pump rotor system according to claim 1, wherein the step S4 is implemented by the following steps:
step S4.1: energy value EmnjWave form factor SmnjPulse index ImnjMargin coefficient LmnjCrest factor CmnjKurtosis value KmnjA fault signature matrix Z of (mn) xj is formed,
Z＝[zE，ZS，ZI，ZL，ZC，ZK，]
step S4.2: normalizing the Z data of the feature matrix; adopting maximum-minimum standardization to perform linear transformation on matrix data and map the original value to an interval [0,1] through maximum-minimum standardization;
wherein x' is the new data after normalization; x is the raw data; minA is the minimum value in the data set; maxA is the maximum value in the data set;
step S4.3: and dividing the feature matrix after the normalization processing into a training sample and a test sample.
3. The method for diagnosing the fault of the centrifugal pump rotor system according to claim 1, wherein the step S5 is implemented by the following steps:
step S5.1: training the limit learning machine model by using the training sample;
step S5.2: and optimizing the extreme learning machine model by adopting a kernel function, wherein the optimized training model becomes a kernel extreme learning machine.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010192412.5A CN111220373A (en) | 2020-03-18 | 2020-03-18 | Method for diagnosing faults of centrifugal pump rotor system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010192412.5A CN111220373A (en) | 2020-03-18 | 2020-03-18 | Method for diagnosing faults of centrifugal pump rotor system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111220373A true CN111220373A (en) | 2020-06-02 |
Family
ID=70828501
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010192412.5A Pending CN111220373A (en) | 2020-03-18 | 2020-03-18 | Method for diagnosing faults of centrifugal pump rotor system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111220373A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112098094A (en) * | 2020-09-27 | 2020-12-18 | 上海数深智能科技有限公司 | Method for diagnosing fault vibration of low-speed heavy-load bearing |
| CN112196784A (en) * | 2020-09-18 | 2021-01-08 | 昆明理工大学 | ELM-based high-pressure diaphragm pump health state estimation system and method |
| CN112231624A (en) * | 2020-09-16 | 2021-01-15 | 中电电气(江苏)变压器制造有限公司 | Real-time evaluation system for short-circuit resistance of multi-transformer winding based on Internet of things |
| CN112798280A (en) * | 2021-02-05 | 2021-05-14 | 山东大学 | Rolling bearing fault diagnosis method and system |
| CN113109050A (en) * | 2021-03-18 | 2021-07-13 | 重庆大学 | Rolling bearing weak fault diagnosis method based on cyclic pulse |
| CN113339280A (en) * | 2021-06-10 | 2021-09-03 | 中国海洋石油集团有限公司 | Offshore centrifugal pump fault diagnosis method and system |
| CN115539378A (en) * | 2022-11-23 | 2022-12-30 | 中汽信息科技(天津)有限公司 | Fault diagnosis method, device and medium for hydraulic gear pump of automobile production line |
| CN115683687A (en) * | 2023-01-03 | 2023-02-03 | 成都大汇物联科技有限公司 | Dynamic and static rub-impact fault diagnosis method for hydroelectric mechanical equipment |
| WO2023123593A1 (en) * | 2021-12-30 | 2023-07-06 | 浙大城市学院 | Variational mode decomposition and residual network-based aviation bearing fault diagnosis method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1920511A (en) * | 2006-08-01 | 2007-02-28 | 东北电力大学 | Fusion diagnosing method of centrifugal pump vibration accidents and vibration signals sampling device |
| CN108241873A (en) * | 2018-01-10 | 2018-07-03 | 中国水利水电科学研究院 | A kind of intelligent failure diagnosis method towards pumping plant main equipment |
| EP3456623A1 (en) * | 2016-06-09 | 2019-03-20 | Hyundai Electric & Energy Systems Co., Ltd. | Method for collecting vessel data using vessel data model, device for collecting vessel data, and vessel comprising same |
| CN109829402A (en) * | 2019-01-21 | 2019-05-31 | 福州大学 | Different operating condition lower bearing degree of injury diagnostic methods based on GS-SVM |
| CN110286286A (en) * | 2019-05-30 | 2019-09-27 | 昆明理工大学 | A kind of VSC-HVDC converter station fault identification device and method based on VMD-ELM |
| CN110515364A (en) * | 2019-07-15 | 2019-11-29 | 北京工业大学 | A kind of cutting-tool wear state detection method based on variation mode decomposition and LS-SVM |
-
2020
- 2020-03-18 CN CN202010192412.5A patent/CN111220373A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1920511A (en) * | 2006-08-01 | 2007-02-28 | 东北电力大学 | Fusion diagnosing method of centrifugal pump vibration accidents and vibration signals sampling device |
| EP3456623A1 (en) * | 2016-06-09 | 2019-03-20 | Hyundai Electric & Energy Systems Co., Ltd. | Method for collecting vessel data using vessel data model, device for collecting vessel data, and vessel comprising same |
| CN108241873A (en) * | 2018-01-10 | 2018-07-03 | 中国水利水电科学研究院 | A kind of intelligent failure diagnosis method towards pumping plant main equipment |
| CN109829402A (en) * | 2019-01-21 | 2019-05-31 | 福州大学 | Different operating condition lower bearing degree of injury diagnostic methods based on GS-SVM |
| CN110286286A (en) * | 2019-05-30 | 2019-09-27 | 昆明理工大学 | A kind of VSC-HVDC converter station fault identification device and method based on VMD-ELM |
| CN110515364A (en) * | 2019-07-15 | 2019-11-29 | 北京工业大学 | A kind of cutting-tool wear state detection method based on variation mode decomposition and LS-SVM |
Non-Patent Citations (1)
| Title |
|---|
| 冯培燕等: "基于改进变分模态分解排列熵和极限学习机的汽轮发电机转子故障诊断方法", 《电机与控制应用》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112231624A (en) * | 2020-09-16 | 2021-01-15 | 中电电气(江苏)变压器制造有限公司 | Real-time evaluation system for short-circuit resistance of multi-transformer winding based on Internet of things |
| CN112231624B (en) * | 2020-09-16 | 2024-03-26 | 中电电气(江苏)变压器制造有限公司 | Real-time evaluation system for short-circuit resistance of multi-transformer winding based on Internet of things |
| CN112196784A (en) * | 2020-09-18 | 2021-01-08 | 昆明理工大学 | ELM-based high-pressure diaphragm pump health state estimation system and method |
| CN112196784B (en) * | 2020-09-18 | 2021-11-26 | 昆明理工大学 | ELM-based high-pressure diaphragm pump health state estimation system and method |
| CN112098094A (en) * | 2020-09-27 | 2020-12-18 | 上海数深智能科技有限公司 | Method for diagnosing fault vibration of low-speed heavy-load bearing |
| CN112798280A (en) * | 2021-02-05 | 2021-05-14 | 山东大学 | Rolling bearing fault diagnosis method and system |
| CN112798280B (en) * | 2021-02-05 | 2022-01-04 | 山东大学 | Rolling bearing fault diagnosis method and system |
| CN113109050A (en) * | 2021-03-18 | 2021-07-13 | 重庆大学 | Rolling bearing weak fault diagnosis method based on cyclic pulse |
| CN113339280A (en) * | 2021-06-10 | 2021-09-03 | 中国海洋石油集团有限公司 | Offshore centrifugal pump fault diagnosis method and system |
| WO2023123593A1 (en) * | 2021-12-30 | 2023-07-06 | 浙大城市学院 | Variational mode decomposition and residual network-based aviation bearing fault diagnosis method |
| CN115539378A (en) * | 2022-11-23 | 2022-12-30 | 中汽信息科技(天津)有限公司 | Fault diagnosis method, device and medium for hydraulic gear pump of automobile production line |
| CN115683687A (en) * | 2023-01-03 | 2023-02-03 | 成都大汇物联科技有限公司 | Dynamic and static rub-impact fault diagnosis method for hydroelectric mechanical equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111220373A (en) | Method for diagnosing faults of centrifugal pump rotor system | |
| CN108171263B (en) | Based on the Fault Diagnosis of Roller Bearings for improving variation mode decomposition and extreme learning machine | |
| Wu et al. | Induction machine fault detection using SOM-based RBF neural networks | |
| CN108731923B (en) | Fault detection method and device for rotary mechanical equipment | |
| Xiang et al. | Fault diagnosis of rolling bearing under fluctuating speed and variable load based on TCO spectrum and stacking auto-encoder | |
| CN112418013B (en) | Complex working condition bearing fault diagnosis method based on meta-learning under small sample | |
| CN110849627A (en) | Width migration learning network and rolling bearing fault diagnosis method based on same | |
| CN108804740B (en) | Long-distance pipeline pressure monitoring method based on integrated improved ICA-KRR algorithm | |
| CN113537328B (en) | Rotary machine fault diagnosis method and device based on deep learning | |
| CN112257530A (en) | Rolling bearing fault diagnosis method based on blind signal separation and support vector machine | |
| CN111459144A (en) | Airplane flight control system fault prediction method based on deep cycle neural network | |
| Omoregbee et al. | Fault detection in roller bearing operating at low speed and varying loads using Bayesian robust new hidden Markov model | |
| Han et al. | An intelligent fault diagnosis method of variable condition gearbox based on improved DBN combined with WPEE and MPE | |
| CN111695452A (en) | Parallel reactor internal aging degree evaluation method based on RBF neural network | |
| CN114354184A (en) | Deep learning-based method and device for establishing health early warning model of main shaft of large-scale rotating equipment | |
| CN114548295A (en) | Bearing fault classification system and method based on multi-scale domain adaptive network | |
| CN113627539A (en) | Rolling bearing composite fault diagnosis method and system, storage medium and computing equipment | |
| Xu et al. | Constructing a health indicator for bearing degradation assessment via an unsupervised and enhanced stacked autoencoder | |
| CN113758708B (en) | Rolling bearing signal frequency domain fault diagnosis method based on L1 norm and group norm constraint | |
| TWI827341B (en) | Application of sample distribution transfer to improve unsupervised learning method for rotating machinery diagnosis | |
| CN112215489A (en) | Industrial equipment anomaly detection method | |
| Wang et al. | Multi-scale convolutional neural network fault diagnosis based on attention mechanism | |
| CN115095535B (en) | Industrial pump operation multi-parameter detector | |
| CN107918828B (en) | Water feeding pump fault detection method based on minimum sufficient statistic model analysis | |
| CN117291094B (en) | Unmanned aerial vehicle power system parameter anomaly detection method based on associated parameter mapping |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200602 |
|
| RJ01 | Rejection of invention patent application after publication |