Intelligent Fault Diagnosis for Rotary Machinery Using Transferable Convolutional Neural Network

TLDR

Deep neural networks achieve competitive results in mechanical fault diagnosis, yet their training demands high computing power and is hampered by limited data. This paper proposes a transferable convolutional neural network to enhance learning for target fault diagnosis tasks. A one‑dimensional CNN is pretrained on large source datasets and then fine‑tuned on target tasks through transfer learning, with the study examining the impact of transfer layers and training sample size across four case studies. The proposed method demonstrates superior classification performance compared with other algorithms.

Abstract

Deep neural networks present very competitive results in mechanical fault diagnosis. However, training deep models require high computing power while the performance of deep architectures in extracting discriminative features for decision making often suffers from the lack of sufficient training data. In this paper, a transferable convolutional neural network (CNN) is proposed to improve the learning of target tasks. First, a one-dimensional CNN is constructed and pretrained based on large source task datasets. Then a transfer learning strategy is adopted to train a deep model on target tasks by reusing the pretrained network. Thus, the proposed method not only utilizes the learning power of deep network but also leverages the prior knowledge from the source task. Four case studies are considered and the effects of transfer layers and training sample size on classification effectiveness are investigated. Results show that the proposed method exhibits better performance compared with other algorithms.