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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Recent transfer learning (TL) approaches in industrial intelligent faultdiagnosis (FD) mostly follow the "pre-train and fine-tuning" paradigm toaddress data drift, which emerges from variable working conditions. However, wefind that this approach is prone to the phenomenon known as catastrophicforgetting. Furthermore, performing frequent models fine-tuning on theresource-constrained edge nodes can be computationally expensive andunnecessary, given the excellent transferability demonstrated by existingmodels. In this work, we propose the Drift-Aware Weight Consolidation (DAWC), amethod optimized for edge deployments, mitigating the challenges posed byfrequent data drift in the industrial Internet of Things (IIoT). DAWCefficiently manages multiple data drift scenarios, minimizing the need forconstant model fine-tuning on edge devices, thereby conserving computationalresources. By detecting drift using classifier confidence and estimatingparameter importance with the Fisher Information Matrix, a tool that measuresparameter sensitivity in probabilistic models, we introduce a drift detectionmodule and a continual learning module to gradually equip the FD model withpowerful generalization capabilities. Experimental results demonstrate that ourproposed DAWC achieves superior performance compared to existing techniqueswhile also ensuring compatibility with edge computing constraints.Additionally, we have developed a comprehensive diagnosis and visualizationplatform.

EdgeFD: An Edge-Friendly Drift-Aware Fault Diagnosis System for Industrial IoT