Rolling bearing fault feature learning using improved convolutional deep belief network with compressed sensing

Shao, Haidong; Jiang, Hongkai; Zhang, Haizhou; Duan, Wenjing; Tan, Liang; Wu, Shuaipeng

doi:10.1016/j.ymssp.2017.08.002

Cited by 330 publications

(128 citation statements)

References 51 publications

Supporting

Mentioning

128

Contrasting

Order By: Relevance

“…b denotes the batch size, and λ is the learning rate. Every class center is initialized as the "batch class center" in the first iteration, and it is updated according to (5) and 6for the next batch of samples in each iteration.…”

Section: Shock and Vibrationmentioning

confidence: 99%

“…Traditional machine learning techniques, especially deep learning, have recently made great achievements in the data-driven fault diagnosis field [1][2][3][4][5][6]. Most machine learning methods assume that the training data (source domain) and test data (target domain) must be in the same working condition and have the same distribution and feature space.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Domain Adaptation Model for Bearing Fault Diagnosis with Domain Alignment and Discriminative Feature Learning

Liu

2020

Shock and Vibration

View full text Add to dashboard Cite

Deep learning techniques have been widely used to achieve promising results for fault diagnosis. In many real-world fault diagnosis applications, labeled training data (source domain) and unlabeled test data (target domain) have different distributions due to the frequent changes of working conditions, leading to performance degradation. This study proposes an end-to-end unsupervised domain adaptation bearing fault diagnosis model that combines domain alignment and discriminative feature learning on the basis of a 1D convolutional neural network. Joint training with classification loss, center-based discriminative loss, and correlation alignment loss between the two domains can adapt learned representations in the source domain for application to the target domain. Such joint training can also guarantee domain-invariant features with good intraclass compactness and interclass separability. Meanwhile, the extracted features can efficiently improve the cross-domain testing performance. Experimental results on the Case Western Reserve University bearing datasets confirm the superiority of the proposed method over many existing methods.

show abstract

Section: Shock and Vibrationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Deep Domain Adaptation Model for Bearing Fault Diagnosis with Domain Alignment and Discriminative Feature Learning

Liu

2020

Shock and Vibration

View full text Add to dashboard Cite

show abstract

“…As this method is computationally expensive, hence it is not suitable for the purpose targeted in this paper. Another method proposed by Shao et al, is using the concept of convolutional deep belief network along with Gaussian visible units for learning the features from compressive measurements [32]. Their method also increases the computational burden and hence, costs power.…”

Section: Background and Related Workmentioning

confidence: 99%

A Machine Condition Monitoring Framework Using Compressed Signal Processing

Rani

Dhok

Deshmukh

2020

Sensors

View full text Add to dashboard Cite

The vibration monitoring of ball bearings of a rotating machinery is a crucial aspect for smooth functioning and sustainability of plants. The wireless vibration monitoring using conventional Nyquist sampling techniques is costly in terms of power consumption, as it generates lots of data that need to be processed. To overcome this issue, compressive sensing (CS) can be employed, which directly acquires the signal in compressed form and hence reduces power consumption. The compressive measurements so generated can easily be transmitted to the base station and the original signal can be recovered there using CS reconstruction algorithms to diagnose the faults. However, the CS reconstruction is very costly in terms of computational time and power. Hence, this conventional CS framework is not suitable for diagnosing the machinery faults in real time. In this paper, a bearing condition monitoring framework is presented based on compressed signal processing (CSP). The CSP is a newer research area of CS, in which inference problems are solved without reconstructing the original signal back from compressive measurements. By omitting the reconstruction efforts, the proposed method significantly improves the time and power cost. This leads to faster processing of compressive measurements for solving the required inference problems for machinery condition monitoring. This gives a way to diagnose the machinery faults in real-time. A comparison of proposed scheme with the conventional method shows that the proposed scheme lowers the computational efforts while simultaneously achieving the comparable fault classification accuracy.

show abstract

“…Jia et al [25] used the normalized sparse AEs to constitute local connection network, and the model can learn to avoid similar, repeated features and overcome the problem of feature change. Shao et al [26] proposed an improved convolution deep belief network method based on compressed sensing technology, this method used compressed data as the input of the model and obtained less time consumption of the fault diagnosis. A novel cross-domain fault diagnosis method was proposed by Li et al [27] whereby multiple deep generative neural networks were employed to generate corresponding-domain fake samples, and faults in different domains could be discriminated well.…”

Section: Introductionmentioning

confidence: 99%

A Novel Intelligent Fault Diagnosis Method for Rolling Bearing Based on Integrated Weight Strategy Features Learning

Ouyang

Chen

et al. 2020

Sensors

View full text Add to dashboard Cite

Intelligent methods have long been researched in fault diagnosis. Traditionally, feature extraction and fault classification are separated, and this process is not completely intelligent. In addition, most traditional intelligent methods use an individual model, which cannot extract the discriminate features when the machines work in a complex condition. To overcome the shortcomings of traditional intelligent fault diagnosis methods, in this paper, an intelligent bearing fault diagnosis method based on ensemble sparse auto-encoders was proposed. Three different sparse auto-encoders were used as the main architecture. To improve the robustness and stability, a novel weight strategy based on distance metric and standard deviation metric was employed to assign the weights of three sparse auto-encodes. Softmax classifier is used to classify the fault types of integrated features. The effectiveness of the proposed method is validated with extensive experiments, and comparisons with the related methods and researches on the widely-used motor bearing dataset verify the superiority of the proposed method. The results show that the testing accuracy and the standard deviation are 99.71% and 0.05%.

show abstract

Rolling bearing fault feature learning using improved convolutional deep belief network with compressed sensing

Cited by 330 publications

References 51 publications

Deep Domain Adaptation Model for Bearing Fault Diagnosis with Domain Alignment and Discriminative Feature Learning

Deep Domain Adaptation Model for Bearing Fault Diagnosis with Domain Alignment and Discriminative Feature Learning

A Machine Condition Monitoring Framework Using Compressed Signal Processing

A Novel Intelligent Fault Diagnosis Method for Rolling Bearing Based on Integrated Weight Strategy Features Learning

Contact Info

Product

Resources

About