A deep reinforcement transfer convolutional neural network for rolling bearing fault diagnosis. (October 2022)

• Record Type: Journal Article
• Authors: Wu, Zhenghong; Jiang, Hongkai; Liu, Shaowei; Wang, Ruixin
• Is Part Of: ISA transactions. Volume 129(2022)Part B
• Page Start: 505
• Page End: 524
• Publication Date: 2022-10
• Subjects: Rolling bearing fault diagnosis -- Deep reinforcement transfer convolution neural network -- Intelligent diagnosis agent -- Parameter transfer learning -- Deep Q-network; Engineering instruments -- Periodicals; Engineering instruments; Periodicals; Electronic journals; 629.805
• DOI: 10.1016/j.isatra.2022.02.032
• Languages: English
• ISSNs: 0019-0578

Abstract

Abstract: Deep neural networks highly depend on substantial labeled samples when identifying bearing fault. However, in some practical situations, it is very difficult to collect sufficient labeled samples, which limits the application of deep neural networks in practical engineering. Therefore, how to use limited labeled samples to complete fault diagnosis tasks is an urgent problem. In this paper, a deep reinforcement transfer convolutional neural network (DRTCNN) is developed to tackle the problem. Firstly, an intelligent diagnosis agent constructed by a convolutional neural network is trained to obtain maximum long-term cumulative rewards, which is characterized by the ability to autonomously learn the latent relationship between fault samples and corresponding labels. Secondly, the parameter transfer learning method is utilized to establish a target task agent of DRTCNN. Finally, limited labeled target domain fault samples and the training mechanism of deep Q-network are employed to train the target task agent for performing target diagnosis tasks. Two diagnosis cases are conducted to verify the effectiveness of the proposed method when only limited labeled target domain fault samples are available. Highlights: DRTCNN for achieving target diagnosis tasks is developed. A diagnosis agent for learning the relationship of samples and labels is designed. Parameter transfer learning is used to establish a target task agent of DRTCNN. DQN training mechanism is employed to train the target task agent of DRTCNN.