Article
Automation & Control Systems
Saibo Xing, Yaguo Lei, Shuhui Wang, Feng Jia
Summary: DIDBN is a deep learning model that directly learns distribution-invariant features from raw vibration data, achieving higher diagnosis accuracies in fault recognition. By utilizing a locally connected RBM and MDM-RBM layer, DIDBN is able to capture features with close distributions under varying working conditions.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2021)
Article
Computer Science, Artificial Intelligence
Jie Yang, Weimin Bao, Yanming Liu, Xiaoping Li
Summary: This paper proposes a joint class metric and sparse representation regularized deep belief network (J-DBN) method for intelligent fault diagnosis of rotary equipment. The proposed method can extract data features, optimize feature distances, and generate sparse features. Experimental results demonstrate that this method significantly enhances feature extraction capability and has higher diagnostic accuracy.
INTERNATIONAL JOURNAL OF INTELLIGENT SYSTEMS
(2022)
Article
Engineering, Electrical & Electronic
Vikas Singh, Nishchal K. Verma
Summary: In recent years, intelligent condition-based monitoring of rotary machinery systems has become a major research focus. Using mRMR and deep learning models can improve fault diagnostics performance by reducing data redundancy and decreasing data dependency for training the model. The proposed frameworks show better diagnostic accuracy and faster processing of data with many features.
IEEE SENSORS JOURNAL
(2021)
Review
Engineering, Electrical & Electronic
Xiaohan Chen, Rui Yang, Yihao Xue, Mengjie Huang, Roberto Ferrero, Zidong Wang
Summary: Traditional deep learning-based bearing fault diagnosis approaches assume that the training and test data follow the same distribution. However, this assumption is not always true for practical bearing data, leading to a decline in fault diagnosis performance. To overcome this, deep transfer learning methods have gained attention by transferring knowledge from other data or models. This review provides a comprehensive overview of deep transfer learning-based bearing fault diagnosis approaches since 2016, proposing a novel taxonomy and discussing challenges and opportunities in the field.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2023)
Article
Engineering, Multidisciplinary
Jiahui Tang, Jimei Wu, Bingbing Hu, Jie Liu
Summary: The proposed method for bearing fault diagnosis based on fault feature region detection utilizes a deep belief network to train the proposed regions with fault features, achieving a prediction accuracy of over 80% for compound faults.
Article
Automation & Control Systems
Mengqi Miao, Jianbo Yu
Summary: In this paper, a sparse representation network (SRNet) is developed to extract impulses from collected signals and used for machinery fault recognition. A convolutional sparse graph is introduced in a sparse representation layer to suppress noise and preserve impulsive characteristics of signals, improving the feature extraction capacity of SRNet. A selective residual learning method is also developed to effectively optimize gradient propagation and further enhance the feature learning performance of SRNet. The feature learning and fault classification capacity of SRNet is evaluated on two gearbox cases, demonstrating its effectiveness compared with other deep neural networks.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2023)
Article
Green & Sustainable Science & Technology
Pengcheng Li, Burkay Anduv, Xu Zhu, Xinqiao Jin, Zhimin Du
Summary: This study collected experimental data of normal operation and refrigerant undercharge fault of chillers in an industrial park, and evaluated the fault sensitivity characteristics of the experimental data. Based on expert knowledge, 23 features were selected as the input of the model. Then, a hybrid model of refrigerant undercharge fault diagnosis for chillers using Deep Belief Network (DBN) enhanced Extreme Learning Machine (ELM) was proposed, and its parameters were optimized by Particle Swarm Optimization (PSO). The experimental results showed that the hybrid model can maintain a high diagnosis accuracy with a small number of training data sets, and it outperforms other models in terms of accuracy and robustness.
SUSTAINABLE ENERGY TECHNOLOGIES AND ASSESSMENTS
(2023)
Article
Automation & Control Systems
Ying Zhang, Jinchen Ji
Summary: A novel method, MI-CDBN, is proposed for fault diagnosis of RCs, utilizing transfer path analysis and multimodal data isolation. Experimental results demonstrate that the proposed method outperforms many other state-of-the-art methods in fault diagnosis of RCs.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2021)
Article
Engineering, Electrical & Electronic
Zhijian Wang, Jie Cui, Wenan Cai, Yanfeng Li
Summary: This article proposes a multidiscriminator deep weighted adversarial network (MDWAN) method for addressing the issue of inconsistent label space between source and target domains in deep transfer learning. By introducing a weight function, this method enhances the positive transfer effect.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Automation & Control Systems
Meidi Sun, Hui Wang, Ping Liu, Shoudao Huang, Pan Wang, Jinhao Meng
Summary: The article proposes a stack autoencoder transfer learning algorithm based on class separation and domain fusion to address issues in intelligent fault diagnosis. The effectiveness of the algorithm is verified through mutual transfer of datasets, showing that it can achieve high accuracy even without labeled fault data in new machines.
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS
(2022)
Article
Automation & Control Systems
Yi Qin, Quan Qian, Yi Wang, Jianghong Zhou
Summary: This study proposes a novel domain adaptation mechanism, called Intermediate Distribution Alignment (IDA), to address the issue of dynamically changing aligning targets in existing mechanisms. By building a feature extractor and utilizing KL divergence, IDA can align the prior distributions of two domains and has been successfully applied to fault transfer diagnosis with better performance compared to typical mechanisms.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2022)
Article
Computer Science, Artificial Intelligence
Huixin Tian, Qiangqiang Xu
Summary: A fault diagnosis method based on spatio-temporal features fusion using deep belief network (STF-DBN) is proposed in this study, which comprehensively processes multi-source signal features of reciprocating compressors. Experimental results confirm the effectiveness of the method in fault detection and early warning.
JOURNAL OF INTELLIGENT MANUFACTURING
(2022)
Article
Engineering, Mechanical
Na Lu, Tao Yin
Summary: This study presents a novel two-stage transferable common feature space mining method called CFCNet for fault diagnosis. By learning and comparing common features and unique features, CFCNet efficiently diagnoses different faults and balances the training progress with few-shot learning strategy. Extensive experiments have verified the superior performance of the proposed method.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2021)
Article
Automation & Control Systems
Jipu Li, Ruyi Huang, Guolin He, Yixiao Liao, Zhen Wang, Weihua Li
Summary: Research on intelligent fault diagnosis based on deep transfer learning has led to the proposal of a two-stage transfer adversarial network for rotating machinery, which can effectively separate new fault types and recognize the quantity. Experimental results indicate that the proposed scheme can address fault diagnosis transfer tasks with multiple new faults in the target domain.
IEEE-ASME TRANSACTIONS ON MECHATRONICS
(2021)
Article
Automation & Control Systems
Chao Zhao, Guokai Liu, Weiming Shen
Summary: This paper proposes a balanced and weighted alignment network for partial transfer fault diagnosis. The method augments the target domain to achieve class balance and reduces marginal distribution discrepancy through a weighted adversarial alignment. Experimental results demonstrate its promising performance on two test rigs.
Article
Energy & Fuels
Fei Mei, Yong Ren, Qingliang Wu, Chenyu Zhang, Yi Pan, Haoyuan Sha, Jianyong Zheng
Article
Energy & Fuels
Danqi Li, Fei Mei, Chenyu Zhang, Haoyuan Sha, Jianyong Zheng
Article
Energy & Fuels
Huiyu Miao, Fei Mei, Yun Yang, Hongfei Chen, Jianyong Zheng
Article
Energy & Fuels
Haoyuan Sha, Fei Mei, Chenyu Zhang, Yi Pan, Jianyong Zheng
Article
Chemistry, Multidisciplinary
Fei Mei, Qingliang Wu, Tian Shi, Jixiang Lu, Yi Pan, Jianyong Zheng
APPLIED SCIENCES-BASEL
(2019)
Article
Chemistry, Multidisciplinary
Deyang Yin, Fei Mei, Jianyong Zheng
APPLIED SCIENCES-BASEL
(2019)
Article
Engineering, Electrical & Electronic
Kedong Zhu, Ning Lu, Jianyong Zheng, Guoqiang Sun, Fei Mei
ELECTRIC POWER SYSTEMS RESEARCH
(2019)
Article
Energy & Fuels
Cheng Zhou, Jianyong Zheng, Sai Liu, Yu Liu, Fei Mei, Yi Pan, Tian Shi, Jianzhang Wu
Article
Energy & Fuels
Tian Shi, Fei Mei, Jixiang Lu, Jinjun Lu, Yi Pan, Cheng Zhou, Jianzhang Wu, Jianyong Zheng
Article
Energy & Fuels
Jiaqi Gu, Fei Mei, Jixiang Lu, Jinjun Lu, Jingcheng Chen, Xinmin Zhang, Limin Li
Article
Energy & Fuels
Xuan Li, Jianyong Zheng, Fei Mei, Haoyuan Sha, Danqi Li
Article
Computer Science, Information Systems
Deyang Yin, Fei Mei, Jianyong Zheng, Weiguo He
Article
Computer Science, Information Systems
Fei Mei, Haoyuan Sha
Article
Computer Science, Information Systems
Haoyuan Sha, Fei Mei, Chenyu Zhang, Yi Pan, Jianyong Zheng, Taoran Li