English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The application of unsupervised domain adaptation (UDA)-based fault diagnosismethods has shown significant efficacy in industrial settings, facilitating thetransfer of operational experience and fault signatures between differentoperating conditions, different units of a fleet or between simulated and realdata. However, in real industrial scenarios, unknown levels and types of noisecan amplify the difficulty of domain alignment, thus severely affecting thediagnostic performance of deep learning models. To address this issue, wepropose an UDA method called Smart Filter-Aided Domain Adversarial NeuralNetwork (SFDANN) for fault diagnosis in noisy industrial scenarios. Theproposed methodology comprises two steps. In the first step, we develop a smartfilter that dynamically enforces similarity between the source and targetdomain data in the time-frequency domain. This is achieved by combining alearnable wavelet packet transform network (LWPT) and a traditional waveletpacket transform module. In the second step, we input the data reconstructed bythe smart filter into a domain adversarial neural network (DANN). To learndomain-invariant and discriminative features, the learnable modules of SFDANNare trained in a unified manner with three objectives: time-frequency featureproximity, domain alignment, and fault classification. We validate theeffectiveness of the proposed SFDANN method based on two fault diagnosis cases:one involving fault diagnosis of bearings in noisy environments and anotherinvolving fault diagnosis of slab tracks in a train-track-bridge couplingvibration system, where the transfer task involves transferring from numericalsimulations to field measurements. Results show that compared to otherrepresentative state of the art UDA methods, SFDANN exhibits superiorperformance and remarkable stability.

Smart filter aided domain adversarial neural network for fault diagnosis in noisy industrial scenarios