Multi-input CNN based vibro-acoustic fusion for accurate fault diagnosis of induction motor

The issue of data-driven cross-domain fault diagnosis for rolling bearings has been effectively addressed through the advancements in domain adaptation (DA) methodologies. However, most existing approaches assume the same set of labels for training data and test data. This assumption often falls short of reality, as new fault types may emerge during the testing phase, resulting in less effective domain adaptation methods based on marginal distribution. To address this issue, this study proposes an open set domain adaptation (OSDA) method based on domain similarity, entropy, confidence and Nuclear-norm 1-Wasserstein discrepancy(DEC-NWD). Within this method, a sample-level transferability criterion is introduced. This criterion quantifies the transferability of target samples and assigns small weights to the unknown class. The complementary nature of entropy and confidence is exploited to improve the discriminability of the network for highly uncertain predictions and to use multiple classifiers to compensate for the possible influence of prediction errors on the confidence.Additionally, the Nuclear-norm 1-Wasserstein discrepancy (NWD) is utilized in this method.It treats the classifier as a discriminator and leverages the predicted discriminative information to maximize the alignment of the common classes between the source and target domains. The proposed method has been validated through extensive experiments conducted on two publicly available bearing datasets.

Section: Related Workmentioning

confidence: 99%

DEC-NWD: an approach for open set domain adaptation in fault diagnosis

Zhao,

Peng,

Xue

et al. 2023

“…Ding and He [25] converted the bearing fault signal into a two-dimensional image by wavelet packet energy, and used the rich fault information contained in the two-dimensional image with the CNN network for fault identification. Similarly, Choudhary et al [26] proposed a combination of non-smooth Gabor transform and constant Q-transform techniques to convert one-dimensional vibration and acoustic signals into time-frequency images, which effectively expressed the non-smooth characteristics of induction motor faults. Although CNN networks have shown great potential in fault diagnosis, the imbalanced amount of datasets between sensors in normal and abnormal conditions always contributes to overfitting problems from models.…”

Section: Introductionmentioning

confidence: 99%

A Bayesian CNN-based fusion framework of sensor fault diagnosis

He,

Zhu,

et al. 2024

Sensors equipped on the high-speed train provide large amounts of data which contributes to its state monitoring. However, it is challenging to distinguish whether the fault originates from the mechanical component or the sensors themselves. The main difficulties lie in the biased amount of normal and fault data as well as the deficiency of multi-source data’s inherent correlation. In this paper, we propose a Bayesian Convolutional neural networks (CNN)-based fusion framework to enhance the ability to identify sensor errors. The framework utilizes wavelet time-frequency maps to extract abnormal features, employs a Bayesian CNN to obtain spatial features from a single sensor, integrates multi-source features via Bidirectional Long Short-Term Memory Network (Bi-LSTM) and enhances the acquired spatial and temporal features using an attention mechanism. The enhanced information finally generated leads to precise identification of the sensor faults. The proposed feature-level fusion framework and the associated attention mechanism facilitate discovering the inherent correlation and filtering of irrelevant information. Results indicate that our proposed method achieves 95.4% in terms of accuracy, which outperforms methods relying on feature extraction with single-source sensors by 7.8%.

“…DL models can automatically extract representative abstract features from raw data using deep architecture design of multiple nonlinear layers and stacked hidden layers [14] and train the model through a large amount of data, effectively solving the limitations of traditional fault diagnosis algorithms. Nowadays, many powerful deep neural network models have been developed, such as convolutional neural networks (CNN) models [15], recurrent neural network (RNN) models [16], transfer learning (TL) network models [17], Transformer network models [18] and so on. Based on the characteristics of rotating machinery signals, researchers have developed DL models suitable for fault diagnosis based on the above model.…”

Section: Introductionmentioning

confidence: 99%

ASG-HOMGAT: a high-order multi-head graph attention network with adaptive small graph structure for rolling bearing fault diagnosis

Ding,

Chen,

Liu

et al. 2024

Traditional Euclidean spatial data processing is difficult to capture the inherent relationships of unstructured data such as bearing vibration signals. Representing vibration signals in graphical form helps to preserve their topologi-cal structure and temporal information. Secondly, most existing graph convolutional network methods are based on large graph structured data, which incurs certain memory overhead when aggregating high-order neighborhood node information and ignores important information between samples in the global graph structure. To address these issues, this paper proposes a high-order multi-head graph attention network based on an adaptive small graph struc-ture (ASG-HOMGAT) for fault diagnosis of rolling bearings. Firstly, the adaptive preprocessing layer is used to adaptively denoise and compress the one-dimensional time-domain vibration signal, generating small rule graph data with topological structure. Then, these small graph structured data samples are input into a higher-order graph neural network, which aggregates features from multiple higher-order neighborhoods to achieve richer feature repre-sentations and fully explore the intrinsic correlation between samples. Finally, these features are aggregated into a reinforced representation of graph nodes through a multi head attention mechanism, and a SoftMax classifier is used for fault classification. The experimental results show that the ASG-HOMGAT method has better performance com-pared to mainstream graph neural network diagnostic models. The code and model will be released at: https://github.com/ding-ss/ASG-HOMGAT.