Attention mechanism in intelligent fault diagnosis of machinery: A review of technique and application

Lv, Haixin; Chen, Jinglong; Zhang, Tianci; Feng, Yong; Liu, Shen

doi:10.1016/j.measurement.2022.111594

Cited by 100 publications

(46 citation statements)

References 174 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To summarize the current research of intelligent FDP, there are a number of outstanding surveys on the topic of intelligent FDP [ 1 , 7 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 ]. They conduct extensive review on existing literature quantitatively and qualitatively from their unique viewpoints, and identify the trends and ideas of FDP methods for different scenarios.…”

Section: Introductionmentioning

confidence: 99%

“…The aforementioned review work provides a very good foundation for the work in this paper. Some surveys concentrate on FDP for specific type of device, e.g., machinery [ 20 , 21 , 22 , 23 , 24 , 27 , 28 ], wind power converter [ 25 ], lithium-ion battery system [ 26 ], while some focus on specific FDP method, e.g., deep domain adaptation [ 21 ], attention mechanism [ 22 ], recurrent neural network (RNN) [ 23 ], etc. Most of these reviews cover the data-driven ML techniques, but few of them give a comprehensive overview of the generic DL techniques used for industrial FDP.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning Techniques in Intelligent Fault Diagnosis and Prognosis for Industrial Systems: A Review

Qiu

Cui

Ping

et al. 2023

Sensors

View full text Add to dashboard Cite

Fault diagnosis and prognosis (FDP) tries to recognize and locate the faults from the captured sensory data, and also predict their failures in advance, which can greatly help to take appropriate actions for maintenance and avoid serious consequences in industrial systems. In recent years, deep learning methods are being widely introduced into FDP due to the powerful feature representation ability, and its rapid development is bringing new opportunities to the promotion of FDP. In order to facilitate the related research, we give a summary of recent advances in deep learning techniques for industrial FDP in this paper. Related concepts and formulations of FDP are firstly given. Seven commonly used deep learning architectures, especially the emerging generative adversarial network, transformer, and graph neural network, are reviewed. Finally, we give insights into the challenges in current applications of deep learning-based methods from four different aspects of imbalanced data, compound fault types, multimodal data fusion, and edge device implementation, and provide possible solutions, respectively. This paper tries to give a comprehensive guideline for further research into the problem of intelligent industrial FDP for the community.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Deep Learning Techniques in Intelligent Fault Diagnosis and Prognosis for Industrial Systems: A Review

Qiu

Cui

Ping

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Therefore, it is necessary to apply intelligent fault diagnosis methods to detect faults in time and identify the root causes. In recent years, attention mechanisms become increasingly popular in the field of fault diagnosis due to the ability to extract global information and achieve efficient resource allocation [2]. Some achievements have been reported to utilize attention mechanisms to improve the performance of models for fault classification.…”

Section: Introductionmentioning

confidence: 99%

SCCAM: Supervised Contrastive Convolutional Attention Mechanism for Ante-hoc Interpretable Fault Diagnosis with Limited Fault Samples

Li¹,

Peng²,

Zhang³

et al. 2023

Preprint

View full text Add to dashboard Cite

In real industrial processes, fault diagnosis methods are required to learn from limited fault samples since the procedures are mainly under normal conditions and the faults rarely occur. Although attention mechanisms have become popular in the field of fault diagnosis, the existing attention-based methods are still unsatisfying for the above practical applications. First, pure attention-based architectures like transformers need a large number of fault samples to offset the lack of inductive biases thus performing poorly under limited fault samples. Moreover, the poor fault classification dilemma further leads to the failure of the existing attention-based methods to identify the root causes. To address the aforementioned issues, we innovatively propose a supervised contrastive convolutional attention mechanism (SCCAM) with ante-hoc interpretability, which solves the root cause analysis problem under limited fault samples for the first time. First, accurate classification results are obtained under limited fault samples. More specifically, we integrate the convolutional neural network (CNN) with attention mechanisms to provide strong intrinsic inductive biases of locality and spatial invariance, thereby strengthening the representational power under limited fault samples. In addition, we ulteriorly enhance the classification capability of the SCCAM method under limited fault samples by employing the supervised contrastive learning (SCL) loss. Second, a novel ante-hoc interpretable attention-based architecture is designed to directly obtain the root causes without expert knowledge. The convolutional block attention module (CBAM) is utilized to directly provide feature contribution behind each prediction thus achieving feature-level explanations. The proposed SCCAM method is tested on a continuous stirred tank heater and the Tennessee Eastman industrial process benchmark. Three common fault diagnosis scenarios are covered, including a balanced scenario for additional verification and two scenarios with limited fault samples (i.e., imbalanced scenario and long-tail scenario). The comprehensive results demonstrate that the proposed SCCAM method can achieve better performance compared with the state-of-the-art methods on fault classification

show abstract

“…CNN in machine learning has fast speed, high accuracy, and good robustness for image classi cation [17][18][19], so it is widely used in fault identi cation and classi cation [20][21][22]. Some scholars use the characteristics of CNN to convert the damage signals in the fault eld into images and input them into CNN for damage identi cation.…”

Section: Introductionmentioning

confidence: 99%

Damage identification of wire rope under noise background via Light- EfficientNetV2 and Magnetic flux leakage image

Huang

Wang

Yang

et al. 2023

Preprint

View full text Add to dashboard Cite

Magnetic flux leakage (MFL) testing, non-destructive testing, can prevent some major accidents of hoist equipment by identifying the damage of wire ropes. However, in harsh working conditions such as mines and oil wells, the inevitable vibration and swing of wire rope will generate noise and interfere with the MFL signal, which makes us difficult to identify the damage. As a classification network, Convolutional neural network (CNN) is positive in recognition accuracy and noise resistance, but it hardly uses in wire rope damage classification. To improve the accuracy of wire rope damage identification under noise background, we propose a method of wire rope damage identification via Light-EfficientNetV2 and MFL image. First, the MFL signal is segmented and rearranged to form the MFL image, and then the image is classified by Light-EfficientNetV2. To improve the classification efficiency, we reduce the layers of EfficientNetV2 to make it lighter. Finally, the availability of this method is proved by the validation set. Compared with four neural networks, the accuracy is the highest. Moreover, as the noise increased, the accuracy of Light-EfficientNetV2 is higher than EfficientNetV2, which has application value in the wire rope damage identification under noise background.

show abstract

Attention mechanism in intelligent fault diagnosis of machinery: A review of technique and application

Cited by 100 publications

References 174 publications

Deep Learning Techniques in Intelligent Fault Diagnosis and Prognosis for Industrial Systems: A Review

Deep Learning Techniques in Intelligent Fault Diagnosis and Prognosis for Industrial Systems: A Review

SCCAM: Supervised Contrastive Convolutional Attention Mechanism for Ante-hoc Interpretable Fault Diagnosis with Limited Fault Samples

Damage identification of wire rope under noise background via Light- EfficientNetV2 and Magnetic flux leakage image

Contact Info

Product

Resources

About