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 incorporation of advanced sensors and machine learning techniques hasenabled modern manufacturing enterprises to perform data-driven in-situ qualitymonitoring based on the sensor data collected in manufacturing processes.However, one critical challenge is that newly presented defect category maymanifest as the manufacturing process continues, resulting in monitoringperformance deterioration of previously trained machine learning models. Hence,there is an increasing need for empowering machine learning model to learncontinually. Among all continual learning methods, memory-based continuallearning has the best performance but faces the constraints of data storagecapacity. To address this issue, this paper develops a novel pseudoreplay-based continual learning by integrating class incremental learning andoversampling-based data generation. Without storing all the data, the developedframework could generate high-quality data representing previous classes totrain machine learning model incrementally when new category anomaly occurs. Inaddition, it could even enhance the monitoring performance since it alsoeffectively improves the data quality. The effectiveness of the proposedframework is validated in an additive manufacturing process, which leveragessupervised classification problem for anomaly detection. The experimentalresults show that the developed method is very promising in detecting novelanomaly while maintaining a good performance on the previous task and brings upmore flexibility in model architecture.

Pseudo Replay-based Class Continual Learning for Online New Category Anomaly Detection in Additive Manufacturing