检索结果-南通市图书馆

Auxiliary generative mutual adversarial networks for class-imbalanced fault diagnosis under small samples

Chinese Journal of Aeronautics 2023年第9期36卷 464-478页

作者： Ranran LI Shunming LI Kun XU Mengjie ZENG Xianglian LI Jianfeng GU Yong CHEN School of Energy and Power Engineering Nanjing University of Aeronautics and AstronauticsNanjing 210016China School of Jincheng Nanjing University of Aeronautics and AstronauticsNanjing 211156China Jiangsu Donghua Testing Technology Co Jingjiang 214500China

The effect of intelligent fault diagnosis of mechanical equipment based on data-driven is often premised on big data and ***,due to the limitation of working environment,operating conditions and equipment status,the fault data collected by mechanical equipment are often small and imbalanced with normal ***,in order to solve the abovementioned dilemma faced by the fault diagnosis of practical mechanical equipment,an auxiliary generative mutual adversarial network(AGMAN)is ***,the generator combined with the auto-encoder(AE)constructs the decoder reconstruction feature loss to assist it to complete the accurate mapping between noise distribution and real data distribution,generate highquality fake samples,supplement the imbalanced dataset to improve the accuracy of small sample class-imbalanced fault ***,the discriminator introduces a structure with unshared dual *** the mutual adversarial between the dual discriminator by setting the scoring criteria that the dual discriminator are completely opposite to the real and fake samples,thus improving the quality and diversity of generated samples to avoid mode ***,the auxiliary generator and the dual discriminator are updated *** auxiliary generator can generate fake samples that deceive both discriminators at the same ***,the dual discriminator cannot give correct scores to the real and fake samples according to their respective scoring criteria,so as to achieve Nash *** three different test-bed datasets for verification,the experimental results show that the proposed method can explicitly generate highquality fake samples,which greatly improves the accuracy of class-unbalanced fault diagnosis under small sample,especially when it is extremely imbalanced,after using this method to supplement fake samples,the fault diagnosis accuracy of DCNN and SAE are relatively big ***,the proposed

关键词： adversarial networks Auto-encoder Class-imbalanced Fault detection Small Samples

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Transfer Learning Based on Joint Feature Matching and adversarial networks

引用

Journal of Shanghai Jiaotong university(Science) 2019年第6期24卷 699-705页

作者： ZHONG Haowen WANG Chao TUO Hongya HU Jian QIAO Lingfeng JING Zhongliang School of Aeronautics and Astronautics Shanghai Jiao Tong UniversityShanghai 200240China

Domain adaptation and adversarial networks are two main approaches for transfer *** adaptation methods match the mean values of source and target domains,which requires a very large batch size during ***,adversarial networks are usually unstable when *** this paper,we propose a joint method of feature matching and adversarial networks to reduce domain discrepancy and mine domaininvariant features from the local and global *** the same time,our method improves the stability of ***,the method is embedded into a unified convolutional neural network that can be easily optimized by gradient *** results show that our joint method can yield the state-of-the-art results on three common public datasets.

关键词： transfer learning adversarial networks feature matching domain-invariant features

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Machinery Fault Diagnosis for Imbalanced Samples via Coupled Generative adversarial networks

Machinery Fault Diagnosis for Imbalanced Samples via Coupled...

引用

第43届中国控制会议

作者： Yatao Liu Xinliang Zhang Yanqi Wang Yitian Zhou Lijie Jia School of Electrical Engineering and Automation Henan Key Laboratory of Intelligent Detection and Control of Coal Mine EquipmentHenan Polytechnic University Anhui Institute of Information Technology Zhoushan Yangwangnaxin Technology Co.Ltd

When applying deep learning networks to the machinery fault diagnosis, it is crucial to train the networks under the sufficient fault samples. To achieve such condition under the imbalance samples, it is essential for the generative networks to learn a complete distribution for the imbalanced samples. Existing data augmentation methods often employ generative models directly to reconstruct samples. Under the limited number of samples, it is a challenge to achieve discriminative features for the fault due to the sample underrepresentation. To address this issue, this paper introduces an innovative approach using coupled generative adversarial networks(CoGAN). The CoGAN framework integrates several systematic enhancements to enhance the robustness of the generative adversarial networks. Firstly, to tackle the imbalance between fault and normal sample distributions,parallel coupled networks are employed to jointly learn their distributions. These networks share weights in the extraction layers of the generator, serving as a representative feature distribution generator for the raw samples. Secondly, a one-sided label smoothing strategy is adopted to mitigate model overfitting in scenarios with limited samples. Finally, historical averaging and gradient penalty are incorporated into the cost function to mitigate adversarial vulnerabilities, ensuring a more stable training performance. Experimental results on the roll bearing dataset from Case Western Reserve University demonstrate that the proposed approach consistently achieves higher accuracy compared to state-of-the-art models across different training sample sizes. This effectively enhances fault diagnosis performance and significantly reduces the dependency on extensive training data.

关键词： deep learning,imbalanced samples fault diagnosis coupled generative model adversarial networks gradient penalty

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：