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

This paper presents a model-free deep reinforcement learning (DRL) approachfor controlling a fiber drawing system. The custom DRL-based control systempredictively regulates fiber diameter and produces a fiber with a desired,constant or non-constant, diameter trajectory, i.e. diameter variation alongthe fiber length. Physical models of the system are not used. The system wastrained and tested on a compact fiber drawing system, which has non-lineardelayed dynamics and stochastic behaviors. For a reference trajectory withrandom step changes, after 1 hour of training, the DRL controller showed thesame root mean squared error (RMSE) as an optimized PI controller; after 3hours of training, it achieved the performance of a quadratic dynamic matrixcontroller (QDMC). While the PI feedback controller showed 3.5 seconds of timelag in a step response, the DRL controller showed less than a second of timelag. Controller performance tests on trajectories not used in the trainingprocess are conducted; for a sine sweep reference trajectory, the DRLcontroller maintained an RMSE under 40 um up to a frequency of 45 mHz, comparedto 25 mHz for QDMC.

Dynamic Control of a Fiber Manufacturing Process using Deep Reinforcement Learning