CN114877925B - Comprehensive energy system sensor fault diagnosis method based on extreme learning machine - Google Patents
Comprehensive energy system sensor fault diagnosis method based on extreme learning machine Download PDFInfo
- Publication number
- CN114877925B CN114877925B CN202210342256.5A CN202210342256A CN114877925B CN 114877925 B CN114877925 B CN 114877925B CN 202210342256 A CN202210342256 A CN 202210342256A CN 114877925 B CN114877925 B CN 114877925B
- Authority
- CN
- China
- Prior art keywords
- fault
- sensor
- energy system
- learning machine
- extreme learning
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D18/00—Testing or calibrating apparatus or arrangements provided for in groups G01D1/00 - G01D15/00
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to a comprehensive energy system sensor fault diagnosis method based on an extreme learning machine, which specifically comprises the following steps: s1, acquiring original signals of a gas turbine sensor in a comprehensive energy system, constructing time sequence representations of various lengths of the original signals, superposing fault signals on the original signals according to fault characteristics, and generating time sequence representations of the fault signals; s2, extracting features of time sequence representation of an original signal and time sequence representation of a fault signal to obtain a feature vector of the fault signal; s3, inputting the feature vectors into an improved multi-kernel extreme learning machine, and training a classifier; s4, inputting the signal to be detected into an integrated model of the feature extractor and the classifier which is trained, and obtaining a fault detection result of the sensor. Compared with the prior art, the invention has the advantages of improving the diagnosis accuracy of the fault type of the sensor signal, shortening the diagnosis time, meeting the requirement of on-line detection in a comprehensive energy system and the like.
Description
Technical Field
The invention relates to the technical field of power engineering, in particular to a comprehensive energy system sensor fault diagnosis method based on an extreme learning machine.
Background
In recent years, in the context of increasing energy demands, improvement of operation efficiency and resource utilization has become a development direction of energy systems. Under this trend, integrated energy systems have been rapidly developed. Integrated energy systems include a variety of primary devices such as gas turbines, steam turbines, compression systems, and energy storage systems. The intelligent of the whole comprehensive energy system and key equipment is beneficial to improving the reliability and availability of the comprehensive energy system, thereby realizing the efficient operation of the system and improving the utilization efficiency of resources. The highly intelligent realization of the integrated energy system needs to rely on real-time monitoring of the operation state of key equipment, accurate operation state information and a reliable control system. Therefore, reliable and effective state monitoring, fault diagnosis and prediction and health management are important for realizing the intelligent comprehensive energy system. The control loops in the integrated energy system and the key equipment thereof are numerous and very complex, so the reliability of the control system is very important for the reliable operation of the integrated energy system and the intellectualization of the integrated energy system. The reliability of the sensor and the accuracy of the output signal are the basis of the reliability of the control system, and early warning of sensor faults is an important way to maintain the reliability of the sensor. In the early years, many studies used empirical methods to determine the cause and location of sensor failures. However, as control technology evolves, control systems become more and more complex. Because of the variety of fault modes and the similarity of signal characteristics among different faults, the diagnosis and the finding of the fault cause are very difficult, and the accuracy and time requirements of the fault diagnosis of the sensor cannot be met by the methods proposed by the prior researches.
The output signals of the devices in the integrated energy system are different from those of other systems. Because of the operating characteristics of the integrated energy system, the output signals of the sensors have inherent characteristics that affect the diagnosis of signal faults, and the influence of these characteristics on the diagnosis of faults should be taken into account when constructing the algorithm. In addition, not only the type of fault but also the degree of fault should be considered in constructing the diagnostic algorithm. In the prior art, the problems are not analyzed in detail, so that the method established based on the prior art is not completely suitable for the fault diagnosis of the sensor of the comprehensive energy system.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a comprehensive energy system sensor fault diagnosis method based on an extreme learning machine, which fully considers the coupling influence caused by signal types, signal inherent characteristics, fault types and fault degrees, overcomes the problems of poor model interpretability and difficult migration, and combines a signal processing method with a machine learning method to effectively realize the comprehensive energy system sensor fault diagnosis.
The aim of the invention can be achieved by the following technical scheme:
the comprehensive energy system sensor fault diagnosis method based on the extreme learning machine specifically comprises the following steps:
s1, acquiring original signals of a gas turbine sensor in a real-running comprehensive energy system, constructing time sequence representations of various lengths of the original signals, superposing fault signals on the original signals according to fault characteristics, and generating time sequence representations of the fault signals;
s2, extracting features of time sequence representation of an original signal and time sequence representation of a fault signal to obtain a feature vector of the fault signal;
s3, inputting the feature vectors into an improved multi-kernel extreme learning machine, and training a classifier;
s4, inputting the signal to be detected into an integrated model of the feature extractor and the classifier which is trained, and obtaining a fault detection result of the sensor.
Types of gas turbine sensors include pressure sensors, temperature sensors, and rotational speed sensors.
The step S1 further includes dividing the acquired raw signals of the plurality of sensors into a training set and a testing set.
Further, the training set and the testing set are provided with a plurality of working condition types, wherein the working condition types comprise start-stop, full load and no-load.
The process of training the classifier by the improved multi-kernel extreme learning machine in the step S3 comprises the following steps:
s31, inputting the training set into the hidden layer to obtain a connection matrix of the input layer and the hidden layer;
s32, acquiring a radial basis function and a group of initial kernel function parameters, establishing a connection matrix and an output matrix between an hidden layer and the kernel function, and comparing the output matrix with a target result matrix;
s33, obtaining optimal kernel function parameters corresponding to the minimum difference between the output matrix and the target result matrix through grid search, and reversely calculating to obtain the corresponding neuron number of the hidden layer, the output matrix and a classification model of the kernel function;
s34, replacing radial basis functions by multiple functions, and repeating the steps S32-S33 to obtain the optimal neuron number and output matrix of each kernel function and a corresponding classification model;
s35, respectively inputting the test set into the classification models of each kernel function, calculating the classification precision of all the classification models, and taking the classification model with the highest classification precision as the final output classifier.
Further, the functions of replacing the radial basis function in step S34 include a linear function, a polygonal function, and a sinusoidal function.
The fault signals superimposed in the step S1 include fault signals of the same type and different fault degrees.
Further, the fault levels include mild, moderate and severe.
The step S2 specifically includes feature extraction by wavelet decomposition, and a 7-dimensional feature vector is established.
Further, the 7-dimensional feature vector includes the mean and variance of the fourth-layer high-frequency component, the fifth-layer high-frequency component, and the low-frequency component of the wavelet decomposition, and the wavelet energy spectrum entropy of the first five layers.
Further, the means and variances of the high-frequency components of the fourth layer, the high-frequency components of the fifth layer and the low-frequency components of the wavelet decomposition are sequentially used as the first 6-dimensional feature vector in the 7-dimensional feature vectors, and the wavelet spectral entropy of the first five layers is used as the 7-dimensional feature vector in the 7-dimensional feature vectors.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention designs an improved technology combining machine learning and signal processing aiming at the sensor faults in the comprehensive energy system by utilizing prior knowledge in space and time, and adds the feature description of fault degree in the superimposed fault signals, thereby effectively improving the accuracy of fault diagnosis and enabling the detection accuracy of the sensor faults in the comprehensive energy system to reach more than 95 percent.
2. The improved multi-kernel extreme learning machine is adopted to train the classifier of the sensor fault, so that the training speed and efficiency of the classifier model are improved, and the memory occupied by the training process is less; and the model diagnosis speed is high, and the requirement of on-line fault diagnosis in the comprehensive energy system can be met.
Drawings
FIG. 1 is a schematic flow chart of the present invention;
fig. 2 is a schematic structural diagram of an improved multi-core extreme learning machine according to the present invention.
Detailed Description
The invention will now be described in detail with reference to the drawings and specific examples. The present embodiment is implemented on the premise of the technical scheme of the present invention, and a detailed implementation manner and a specific operation process are given, but the protection scope of the present invention is not limited to the following examples.
Examples
As shown in fig. 1, the fault diagnosis method for the comprehensive energy system sensor based on the extreme learning machine specifically comprises the following steps:
s1, acquiring original signals of a gas turbine sensor in a real-running comprehensive energy system, constructing time sequence representations of various lengths of the original signals, superposing fault signals on the original signals according to fault characteristics, and generating time sequence representations of the fault signals;
s2, extracting features of time sequence representation of an original signal and time sequence representation of a fault signal to obtain a feature vector of the fault signal;
s3, inputting the feature vectors into an improved multi-kernel extreme learning machine, and training a classifier;
s4, inputting the signal to be detected into an integrated model of the feature extractor and the classifier which is trained, and obtaining a fault detection result of the sensor.
Types of gas turbine sensors include pressure sensors, temperature sensors, and rotational speed sensors.
Step S1 further includes dividing the acquired raw signals of the plurality of sensors into a training set and a testing set.
The training set and the test set are provided with various working condition types, wherein the working condition types comprise start-stop, full load and no-load.
The improved multi-core extreme learning machine in step S3 is specifically a multi-aggregate first learning machine.
The process of training the classifier by the improved multi-kernel extreme learning machine in the step S3 comprises the following steps:
s31, inputting the training set into the hidden layer to obtain a connection matrix of the input layer and the hidden layer;
s32, acquiring a radial basis function and a group of initial kernel function parameters, establishing a connection matrix and an output matrix between an hidden layer and the kernel function, and comparing the output matrix with a target result matrix;
s33, obtaining optimal kernel function parameters corresponding to the minimum difference between the output matrix and the target result matrix through grid search, and reversely calculating to obtain the corresponding neuron number of the hidden layer, the output matrix and a classification model of the kernel function;
s34, replacing radial basis functions by multiple functions, and repeating the steps S32-S33 to obtain the optimal neuron number and output matrix of each kernel function and a corresponding classification model;
s35, respectively inputting the test set into the classification models of each kernel function, calculating the classification precision of all the classification models, and taking the classification model with the highest classification precision as the final output classifier.
The functions replacing the radial basis function in step S34 include a linear function, a polygonal function, and a sinusoidal function.
The kernel extreme learning machine is a single hidden layer feedforward neural network, which consists of an input layer, a hidden layer and an output layer. Let ω be the connection weight matrix between the input layer and the hidden layer, β be the connection weight matrix between the hidden layer and the output layer, and the activation function of the hidden layer neurons be g (x). The relationship between the hidden layer output matrix being H, the output matrix of the network being T and β is:
where x is the input, b is the connection weight, and C is the regularization coefficient.
The kernel function is introduced in the kernel Extreme Learning Machine (ELM) network, and the kernel matrix is expressed as follows:
Ω ELM ＝H T H＝h(x i )h(x j )＝K(x i ,x j )
wherein Ω ELM Is a kernel matrix, K is a kernel function, x i And x j Is an input vector, the kernel extreme learning machine F (x) can be expressed as:
when the traditional kernel extreme learning machine trains the classifier aiming at different inputs, the combination needs to be manually adjusted to find the optimal combination of the kernel function and the hidden layer neuron number. As shown in FIG. 2, the present embodiment improves the process of manually searching the optimal combination, forms a multi-kernel extreme learning machine algorithm, avoids manual searching during training, and saves the time for model training.
The fault signals superimposed in step S1 include fault signals of the same type and different degrees of fault.
The degree of failure includes mild, moderate and severe.
The step S2 specifically includes feature extraction by wavelet decomposition, and 7-dimensional feature vectors are created.
The 7-dimensional feature vector includes the mean and variance of the fourth-layer high-frequency component, the fifth-layer high-frequency component, and the low-frequency component of the wavelet decomposition, and the wavelet energy spectral entropy of the first five layers.
The mean and variance of the high-frequency component of the fourth layer, the high-frequency component of the fifth layer and the low-frequency component of the wavelet decomposition are sequentially used as the first 6-dimensional feature vector in the 7-dimensional feature vectors, and the wavelet energy spectrum entropy of the first five layers is used as the 7-dimensional feature vector in the 7-dimensional feature vectors.
Furthermore, the particular embodiments described herein may vary from one embodiment to another, and the above description is merely illustrative of the structure of the present invention. Equivalent or simple changes of the structure, characteristics and principle of the present invention are included in the protection scope of the present invention. Various modifications or additions to the described embodiments or similar methods may be made by those skilled in the art without departing from the structure of the invention or exceeding the scope of the invention as defined in the accompanying claims.
Claims (6)
1. The comprehensive energy system sensor fault diagnosis method based on the extreme learning machine is characterized by comprising the following steps of:
s1, acquiring original signals of a gas turbine sensor in a real-running comprehensive energy system, constructing time sequence representations of various lengths of the original signals, superposing fault signals on the original signals according to fault characteristics, and generating time sequence representations of the fault signals;
s2, extracting features of time sequence representation of an original signal and time sequence representation of a fault signal to obtain a feature vector of the fault signal;
s3, inputting the feature vectors into an improved multi-kernel extreme learning machine, and training a classifier;
s4, inputting the signal to be detected into an integrated model of the feature extractor and the classifier which is trained, and obtaining a fault detection result of the sensor;
types of the gas turbine sensor include a pressure sensor, a temperature sensor, and a rotational speed sensor;
the step S1 further comprises dividing the acquired original signals of the plurality of sensors into a training set and a testing set, wherein the training set and the testing set are provided with a plurality of working condition types, and the working condition types comprise start and stop, full load and no load;
the process of training the classifier by the improved multi-kernel extreme learning machine in the step S3 comprises the following steps:
s31, inputting the training set into the hidden layer to obtain a connection matrix of the input layer and the hidden layer;
s32, acquiring a radial basis function and a group of initial kernel function parameters, establishing a connection matrix and an output matrix between an hidden layer and the kernel function, and comparing the output matrix with a target result matrix;
s33, obtaining optimal kernel function parameters corresponding to the minimum difference between the output matrix and the target result matrix through grid search, and reversely calculating to obtain the corresponding neuron number of the hidden layer, the output matrix and a classification model of the kernel function;
s34, replacing radial basis functions by multiple functions, and repeating the steps S32-S33 to obtain the optimal neuron number and output matrix of each kernel function and a corresponding classification model;
s35, respectively inputting the test set into the classification models of each kernel function, calculating the classification precision of all the classification models, and taking the classification model with the highest classification precision as the final output classifier.
2. The method for diagnosing a sensor fault of an integrated energy system based on an extreme learning machine according to claim 1, wherein the fault signals superimposed in the step S1 include fault signals of the same type and different fault degrees.
3. The method for diagnosing a fault in an integrated energy system sensor based on an extreme learning machine as recited in claim 2, wherein the degree of the fault signal includes a slight, a medium and a severe degree.
4. The method for diagnosing a sensor fault of an integrated energy system based on an extreme learning machine according to claim 1, wherein the step S2 specifically includes feature extraction by wavelet decomposition, and a 7-dimensional feature vector is established.
5. The method for diagnosing a sensor fault in an integrated energy system based on an extreme learning machine according to claim 4, wherein the 7-dimensional feature vector includes the mean and variance of the fourth-layer high-frequency component, the fifth-layer high-frequency component and the low-frequency component of the wavelet decomposition, and the wavelet spectral entropy of the first five layers.
6. The method for diagnosing the sensor fault of the comprehensive energy system based on the extreme learning machine according to claim 5, wherein the mean and the variance of the fourth-layer high-frequency component, the fifth-layer high-frequency component and the low-frequency component of the wavelet decomposition are sequentially used as the first 6-dimensional feature vector in the 7-dimensional feature vectors, and the wavelet spectral entropy of the first five layers is used as the 7-dimensional feature vector in the 7-dimensional feature vectors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210342256.5A CN114877925B (en) | 2022-03-31 | 2022-03-31 | Comprehensive energy system sensor fault diagnosis method based on extreme learning machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210342256.5A CN114877925B (en) | 2022-03-31 | 2022-03-31 | Comprehensive energy system sensor fault diagnosis method based on extreme learning machine |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114877925A CN114877925A (en) | 2022-08-09 |
| CN114877925B true CN114877925B (en) | 2023-08-22 |
Family
ID=82669523
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210342256.5A Active CN114877925B (en) | 2022-03-31 | 2022-03-31 | Comprehensive energy system sensor fault diagnosis method based on extreme learning machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114877925B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105740619A (en) * | 2016-01-28 | 2016-07-06 | 华南理工大学 | On-line fault diagnosis method of weighted extreme learning machine sewage treatment on the basis of kernel function |
| CN107122860A (en) * | 2017-04-28 | 2017-09-01 | 辽宁工程技术大学 | Bump danger classes Forecasting Methodology based on grid search and extreme learning machine |
| CN111752259A (en) * | 2020-06-02 | 2020-10-09 | 上海交通大学 | Fault identification method and device for gas turbine sensor signal |
| AU2020103811A4 (en) * | 2020-11-18 | 2021-02-11 | Shandong Kuaimai Electronic Technology Co., Ltd. | Method for recognizing human knee motion postures based on extreme learning machine |
| CN112393934A (en) * | 2020-11-20 | 2021-02-23 | 湖南工业大学 | Wind turbine generator fault diagnosis method based on sparse self-coding and extreme learning machine |
| CN112766140A (en) * | 2021-01-15 | 2021-05-07 | 云南电网有限责任公司电力科学研究院 | Transformer fault identification method based on kernel function extreme learning machine |
| CN113255546A (en) * | 2021-06-03 | 2021-08-13 | 成都卡莱博尔信息技术股份有限公司 | Diagnosis method for aircraft system sensor fault |
-
2022
- 2022-03-31 CN CN202210342256.5A patent/CN114877925B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105740619A (en) * | 2016-01-28 | 2016-07-06 | 华南理工大学 | On-line fault diagnosis method of weighted extreme learning machine sewage treatment on the basis of kernel function |
| CN107122860A (en) * | 2017-04-28 | 2017-09-01 | 辽宁工程技术大学 | Bump danger classes Forecasting Methodology based on grid search and extreme learning machine |
| CN111752259A (en) * | 2020-06-02 | 2020-10-09 | 上海交通大学 | Fault identification method and device for gas turbine sensor signal |
| AU2020103811A4 (en) * | 2020-11-18 | 2021-02-11 | Shandong Kuaimai Electronic Technology Co., Ltd. | Method for recognizing human knee motion postures based on extreme learning machine |
| CN112393934A (en) * | 2020-11-20 | 2021-02-23 | 湖南工业大学 | Wind turbine generator fault diagnosis method based on sparse self-coding and extreme learning machine |
| CN112766140A (en) * | 2021-01-15 | 2021-05-07 | 云南电网有限责任公司电力科学研究院 | Transformer fault identification method based on kernel function extreme learning machine |
| CN113255546A (en) * | 2021-06-03 | 2021-08-13 | 成都卡莱博尔信息技术股份有限公司 | Diagnosis method for aircraft system sensor fault |
Non-Patent Citations (1)
| Title |
|---|
| 基于极限学习机的谐波检测研究;王锦国;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ级》(第2期);C042-1156 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114877925A (en) | 2022-08-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110849626B (en) | Self-adaptive sparse compression self-coding rolling bearing fault diagnosis system | |
| Gou et al. | Aeroengine control system sensor fault diagnosis based on CWT and CNN | |
| Zhao et al. | Intelligent fault diagnosis of multichannel motor–rotor system based on multimanifold deep extreme learning machine | |
| Pan et al. | A novel deep learning network via multiscale inner product with locally connected feature extraction for intelligent fault detection | |
| WO2022077901A1 (en) | Bearing failure mode diagnosis method using small sample data sets, and system | |
| Yin et al. | Wasserstein generative adversarial network and convolutional neural network (WG-CNN) for bearing fault diagnosis | |
| CN110929765A (en) | Convolution self-coding fault monitoring method based on batch imaging | |
| CN113420837B (en) | Fault diagnosis method based on multi-source compressed sensing | |
| Zhang et al. | Intelligent machine fault diagnosis using convolutional neural networks and transfer learning | |
| Dong et al. | Design and application of unsupervised convolutional neural networks integrated with deep belief networks for mechanical fault diagnosis | |
| CN111192379A (en) | Comprehensive fault diagnosis method for complete aircraft | |
| CN111783531A (en) | Water turbine set fault diagnosis method based on SDAE-IELM | |
| Afrasiabi et al. | Wind turbine fault diagnosis with generative-temporal convolutional neural network | |
| Zhang et al. | A new deep convolutional domain adaptation network for bearing fault diagnosis under different working conditions | |
| Karamti et al. | A new deep stacked architecture for multi-fault machinery identification with imbalanced samples | |
| CN115587290A (en) | Aero-engine fault diagnosis method based on variational self-coding generation countermeasure network | |
| Ribeiro et al. | Rotating machinery fault diagnosis using similarity-based models | |
| CN115791174B (en) | Rolling bearing abnormality diagnosis method, system, electronic equipment and storage medium | |
| CN114877925B (en) | Comprehensive energy system sensor fault diagnosis method based on extreme learning machine | |
| CN115099264A (en) | Ship part fault diagnosis method and device, computer and computer storage medium | |
| Zhang et al. | Multi‐attribute quantitative bearing fault diagnosis based on convolutional neural network | |
| Hu et al. | An Improved Metalearning Framework to Optimize Bearing Fault Diagnosis under Data Imbalance | |
| Zhu et al. | Compound Fault Diagnosis of Rolling Bearings with Few-shot based on DCGAN-RepLKNet | |
| Xie et al. | Information Fusion-Based Meta-Learning for Few-Shot Fault Diagnosis Under Different Working Conditions | |
| Pan et al. | Generalized Zero-shot Learning Fault Diagnosis Framework Based on Anomaly Detection and Contractive Stacked Autoencoder |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |