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

High-quality mesh generation is the foundation of accurate finite elementanalysis. Due to the vast interior vertices search space and complex initialboundaries, mesh generation for complicated domains requires substantial manualprocessing and has long been considered the most challenging and time-consumingbottleneck of the entire modeling and analysis process. In this paper, wepresent a novel computational framework named ``SRL-assisted AFM" for meshingplanar geometries by combining the advancing front method with neural networksthat select reference vertices and update the front boundary using ``policynetworks." These deep neural networks are trained using a unique pipeline thatcombines supervised learning with reinforcement learning to iteratively improvemesh quality. First, we generate different initial boundaries by randomlysampling points in a square domain and connecting them sequentially. Theseboundaries are used for obtaining input meshes and extracting training datasetsin the supervised learning module. We then iteratively improve thereinforcement learning model performance with reward functions designed forspecial requirements, such as improving the mesh quality and controlling thenumber and distribution of extraordinary points. Our proposed supervisedlearning neural networks achieve an accuracy higher than 98% on predictingcommercial software. The final reinforcement learning neural networksautomatically generate high-quality quadrilateral meshes for complex planardomains with sharp features and boundary layers.

SRL-Assisted AFM: Generating Planar Unstructured Quadrilateral Meshes with Supervised and Reinforcement Learning-Assisted Advancing Front Method