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

Recent progress in artificial intelligence (AI) and high-performancecomputing (HPC) have brought potentially game-changing opportunities inaccelerating reactive flow simulations. In this study, we introduce anopen-source computational fluid dynamics (CFD) framework that integrates thestrengths of machine learning (ML) and graphics processing unit (GPU) todemonstrate their combined capability. Within this framework, all computationaloperations are solely executed on GPU, including ML-accelerated chemistryintegration, fully-implicit solving of PDEs, and computation of thermal andtransport properties, thereby eliminating the CPU-GPU memory copy overhead.Optimisations both within the kernel functions and during the kernel launchprocess are conducted to enhance computational performance. Strategies such asstatic data reorganisation and dynamic data allocation are adopted to reducethe GPU memory footprint. The computational performance is evaluated in twoturbulent flame benchmarks using quasi-DNS and LES modelling, respectively.Remarkably, while maintaining a similar level of accuracy to the conventionalCPU/CVODE-based solver, the GPU/ML-accelerated approach shows an overallspeedup of over two orders of magnitude for both cases. This result highlightsthat high-fidelity turbulent combustion simulation with finite-rate chemistrythat requires normally hundreds of CPUs can now be performed on portabledevices such as laptops with a medium-end GPU.

An integrated framework for accelerating reactive flow simulation using GPU and machine learning models