Comparisons of different deep learning-based methods on fault diagnosis for geared system

Han, Bing; Yang, Xiaohui; Ren, Yafeng; Lan, Wanggui

doi:10.1177/1550147719888169

Cited by 15 publications

(13 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, the DBN has strong nonlinear processing ability and good discriminant ability, so that it has been widely used in the many fields, such as image processing [26], human action identification [27], natural language processing [28]. The DBN model has been preliminarily applied in bearing fault diagnosis [29][30][31] and gear fault diagnosis [29,[32][33][34] since it has been applied to aero-engine structural health identification by scholars [34], however, it has not yet been reported for the identification of fault parameters in double-rotor misalignment.…”

Section: Introductionmentioning

confidence: 99%

Dual-rotor misalignment fault quantitative identification based on DBN and improved D-S evidence theory

Yang

Fanyu

Miao

et al. 2021

Mechanics & Industry

View full text Add to dashboard Cite

Misalignment fault is the main factor that affects the normal running of dual-rotor system. Quantitative identification the misalignment fault is an important way to ensure the safe and stable service of the dual-rotor system, while the identification accuracy of traditional methods is low. Aiming at the above problems, this paper proposed a dual-rotor misalignment fault quantitative identification method based on DBN and D-S evidence theory improved by mutual information measure (MIMD-S). Seven groups experiments were conducted and several vibration signals were collected. By comparing it with the traditional methods D-S, and Pignistic improved D-S (PD-S) evidence theory, the results show that the method proposed in this paper improves the accuracy of the misalignment fault quantitative identification of the dual-rotor, the identification error rate was only 0.36%.

show abstract

Section: Introductionmentioning

confidence: 99%

Dual-rotor misalignment fault quantitative identification based on DBN and improved D-S evidence theory

Yang

Fanyu

Miao

et al. 2021

Mechanics & Industry

View full text Add to dashboard Cite

show abstract

“…The fault in mechanical system is the abnormal state caused by the damage of its component, which induces the mechanical system to be dis‐functional. The fault can be monitored, recognized, and diagnosed by equipment model, running parameter, dynamic response and symptoms 1,2 . Due to the equipment's power frequency vibration, electrical noise, hydraulic system pulsation noise, and noise interference in transmission link modulation, the features in online monitoring system is weak and the feature extraction method is sensitive to the interference factors.…”

Section: Introductionmentioning

confidence: 99%

“…The fault can be monitored, recognized, and diagnosed by equipment model, running parameter, dynamic response and symptoms. 1,2 Due to the equipment's power frequency vibration, electrical noise, hydraulic system pulsation noise, and noise interference in transmission link modulation, the features in online monitoring system is weak and the feature extraction method is sensitive to the interference factors. The solution for mechanical fault diagnosis has become a hot topic in the field of mechanical design.…”

Section: Introductionmentioning

confidence: 99%

Automatic machinery fault detection via using distributed sensor information

Chen

Huo

2021

Internet Technology Letters

View full text Add to dashboard Cite

It is critical in intelligent manufacturing that monitors devices to sustain normal status. Recently, the artificial intelligence‐powered diagnosis is the key in smart monitoring process and has become a hot topic in the engineering field. In previous diagnostic methods, complete failure samples are the premise to activate the intelligent diagnostic model. However, the actual failures of the mechanical diagnostic are far more than that we can obtain under running state in advance. In order solve this issue, this paper adopts Generative Adversarial Networks (GAN) by using simulated data to obtain complete failure samples, which builds a bridge between artificial intelligent model and mechanical fault diagnostic system. In mechanical diagnostic system, we first use finite element simulation to generate the missing fault samples. The generated and existing fault samples are supplemented each other to construct complete failure samples. The complete failure samples are used as training set to train an intelligent model which is used to predict the status for future status of mechanical system. The status of mechanical system is represented as the signal from distributed sensors. The fault diagnostic system is verified on a public mechanical system dataset and the results demonstrate it is superior to previous ones.

show abstract

“…Vibration signal analysis involves separating the collected vibration signal from the fault characteristic signal and identifying the fault in the system by analyzing the separated signal [ 9 ]. In the last few decades, many signal-based fault diagnosis methods have been developed for fault feature identification [ 4 ].…”

Section: Introductionmentioning

confidence: 99%

“…Although the performance of fault diagnosis systems depends on the feature spaces used [ 28 ], it was observed that these feature spaces were used arbitrarily and not combined in such a way to achieve generalization between different conditions, that the bearings may be subjected. For instance, in [ 9 ] while the FFT would give unsatisfactory results with a signal whose frequency components changes with time, the wavelet transform is known to suffer from fixed scale resolution which would affect its real-life applications, and the HHT suffers from instability in its signal decomposition process.…”

Section: Introductionmentioning

confidence: 99%

Health Condition Estimation of Bearings with Multiple Faults by a Composite Learning-Based Approach

Inyang

Petrunin

Jennions

2021

Sensors

View full text Add to dashboard Cite

Bearings are critical components found in most rotating machinery; their health condition is of immense importance to many industries. The varied conditions and environments in which bearings operate make them prone to single and multiple faults. Widespread interest in the improvements of single fault diagnosis meant limited attention was spent on multiple fault diagnosis. However, multiple fault diagnosis poses extra challenges due to the submergence of the weak fault by the strong fault, presence of non-Gaussian noise, coupling of the frequency components, etc. A number of existing convolutional neural network models operate on a distinct feature that is not enough to assure reliable results in the presence of these challenges. In this paper, extended feature sets in three homogenous deep learning models are used for multiple fault diagnosis. This ensures a measure of diversity is introduced to the health management dataset to obtain complementary solutions from the models. The outputs of the models are fused through blending ensemble learning. Experiments using vibration datasets based on bearing multiple faults show an accuracy of 98.54%, with an improvement of 2.74% in the overall effectiveness over the single models. Compared with other technologies, the results show that this approach provides an improved generalized diagnostic capability.

show abstract

Comparisons of different deep learning-based methods on fault diagnosis for geared system

Cited by 15 publications

References 33 publications

Dual-rotor misalignment fault quantitative identification based on DBN and improved D-S evidence theory

Dual-rotor misalignment fault quantitative identification based on DBN and improved D-S evidence theory

Automatic machinery fault detection via using distributed sensor information

Health Condition Estimation of Bearings with Multiple Faults by a Composite Learning-Based Approach

Contact Info

Product

Resources

About