Institutional Repository of Coll Elect & Informat Engn
Method to enhance deep learning fault diagnosis by generating adversarial samples | |
Cao, Jie1,4; Ma, Jialin1; Huang, Dailin1; Yu, Ping2; Wang, Jinhua2; Zheng, Kangjie3 | |
2022-02 | |
发表期刊 | Applied Soft Computing |
ISSN | 1568-4946 |
卷号 | 116 |
摘要 | Modern industrial fields utilize complex mechanical equipment and machinery, which are closely linked, and equipment faults are difficult to express. Therefore, fault diagnosis is important to ensure the safety of complex mechanical equipment in modern industries. Deep learning has achieved excellent results with recent fault diagnosis methods. At present, three common deep learning models (MLP, CNN, and RNN models) can achieve diagnosis rates close to 100% with original fault diagnosis data and a signal-to-noise ratio above 10 dB. However, we found that the diagnostic rate of these three models was completely incorrect when an adversarial sample with a signal-to-noise ratio noise greater than 10 dB was added to the original sample. We propose a GAN-based adversarial signal generative adversarial network (AdvSGAN) in this paper. We conduct experiments on the CWRU dataset and conclude that we can easily obtain adversarial noise and generate training samples through AdvSGAN. With the addition of adversarial data training, the diagnostic rate of the model on these adversarial samples increased from less than 5% to 98.69%, 97.38% and 96.94%. Hence, this method increases the reliability of our deep learning model. © 2021 Elsevier B.V. |
关键词 | Accident prevention Complex networks Convolutional neural networks Fault detection Generative adversarial networks Machinery Recurrent neural networks Signal to noise ratio Adversarial-signal generative adversarial network Convolutional neural network Fault diagnosis method Faults diagnosis Industrial fields Learning models Mechanical equipment Multilayers perceptrons Original sample Three models |
DOI | 10.1016/j.asoc.2021.108385 |
收录类别 | EI ; SCIE |
语种 | 英语 |
WOS研究方向 | Computer Science |
WOS类目 | Computer Science, Artificial Intelligence ; Computer Science, Interdisciplinary Applications |
WOS记录号 | WOS:000768204300009 |
出版者 | Elsevier Ltd |
EI入藏号 | 20220211449083 |
EI主题词 | Failure analysis |
EI分类号 | 716.1 Information Theory and Signal Processing ; 722 Computer Systems and Equipment ; 723.4 Artificial Intelligence ; 914.1 Accidents and Accident Prevention |
来源库 | WOS |
引用统计 | |
文献类型 | 期刊论文 |
条目标识符 | https://ir.lut.edu.cn/handle/2XXMBERH/157945 |
专题 | 电气工程与信息工程学院 |
通讯作者 | Ma, Jialin |
作者单位 | 1.Lanzhou Univ Technol, Coll Comp & Commun, Lanzhou 730050, Peoples R China; 2.Lanzhou Univ Technol, Coll Elect & Informat Engn, Lanzhou 730050, Peoples R China; 3.Dalian Maritime Univ, Informat Sci & Engn, Dalian 116000, Peoples R China; 4.Engn Res Ctr Mfg Informat Gansu Prov, Lanzhou 287, Peoples R China |
第一作者单位 | 计算机与通信学院 |
通讯作者单位 | 计算机与通信学院 |
第一作者的第一单位 | 计算机与通信学院 |
推荐引用方式 GB/T 7714 | Cao, Jie,Ma, Jialin,Huang, Dailin,et al. Method to enhance deep learning fault diagnosis by generating adversarial samples[J]. Applied Soft Computing,2022,116. |
APA | Cao, Jie,Ma, Jialin,Huang, Dailin,Yu, Ping,Wang, Jinhua,&Zheng, Kangjie.(2022).Method to enhance deep learning fault diagnosis by generating adversarial samples.Applied Soft Computing,116. |
MLA | Cao, Jie,et al."Method to enhance deep learning fault diagnosis by generating adversarial samples".Applied Soft Computing 116(2022). |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论