9 September 2022 in IEEE Transactions on Industrial Informatics
Mechanical fault diagnosis is crucial to ensure the safe operations of equipment in intelligent manufacturing systems. Deep learning-based methods have been recently developed for fault diagnosis due to their advantages in feature representation. However, most of these methods fail to learn relations between samples and thus perform poorly without sufficient labeled data. In this article, we propose a new few-shot learning method named dual graph neural network (DGNNet) with residual blocks to address fault diagnosis problems with limited data. First, the residual module learns the feature of samples with image data transferred from original signals. Second, two complete graphs built on the sample features are used to extract the instance-level and distribution-level relations between samples. In particular, an alternate update policy between the instance and distribution graphs integrates the multilevel relations to propagate the label information of a few labeled samples to unlabeled samples. This technique leverages labeled and unlabeled samples to identify unseen faults, encouraging DGNNet competency in fault diagnosis tasks with very few labeled samples. Extensive results on various datasets show that DGNNet achieves excellent performance in supervised fault diagnosis tasks and outperforms baselines by a great margin in semisupervised cases.
13 January 2024 in Expert Systems with Applications
26 November 2019 in IEEE Access
Early fault detection and diagnosis can increase the stability, reliability and safety of manufacturing equipment. It can be used for protection against unforeseen emergencies in manufacturing system. Recently, fault diagnosis (FD) methods based on deep learning (DL) have become a research hotspot for their excellent performance. However, the training process of deep learning (DL) models is time-consuming because of their high computation complexity. Moreover, most of DL-based FD methods have an assumption the distribution of training datasets in the source domain is the same as that of test datasets in the target domain. However, it is impossible in typical real-world manufacturing applications. In order to cope with these two problems, this paper proposes a FD method based on convolutional neural network (CNN) and transfer learning (TL). Firstly, a CNN model based on LeNet-5 is designed to extract fault features from images which is converted from raw signal data by continuous wavelet transform (CWT), then the performance of the CNN model are improved by fine-tuning which is an effective way of TL. The proposed method is conducted on two well-known datasets and the experimental results show that the proposed method can significantly improve the accuracy and efficiency performance a lot compared with the standard CNN model.
11 November 2024 in IEEE Transactions on Industrial Informatics
Industrial time series prediction (ITSP) is critical to the predictive maintenance system of modern industry. However, time-varying conditions and complex industrial processes cause the distribution drift of industrial time series, raising the difficulty of prediction. This article proposes an ITSP model considering distribution information, namely MDEformer. First, the multivariate dynamic embedding (MDE) is designed to provide the property of the channel-binding dynamic distribution awareness. Specifically, a dynamic mode transition and selection module is adopted to exploit dynamic distribution features of time series, and the bidirectional dynamic residual connection integrates dynamic distribution information into embedding vectors to filter distribution change interference. Then, the vanilla Transformer encoder is used to achieve multivariate prediction. Finally, a generative pretraining and fine-tuning strategy is used to enhance the generalization ability in real production scenarios. Extensive results on a real-world zinc smelting dataset illustrate the superiority of MDEformer.
18 April 2022 in Resources Policy
4 November 2021 in The International Journal of Advanced Manufacturing Technology