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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

In this work, we introduce DeepFlame, an open-source C++ platform with thecapabilities of utilising machine learning algorithms and pre-trained models tosolve for reactive flows. We combine the individual strengths of thecomputational fluid dynamics library OpenFOAM, machine learning frameworkTorch, and chemical kinetics program Cantera. The complexity of cross-libraryfunction and data interfacing (the core of DeepFlame) is minimised to achieve asimple and clear workflow for code maintenance, extension and upgrading. As ademonstration, we apply our recent work on deep learning for predictingchemical kinetics (Zhang et al. Combust. Flame vol. 245 pp. 112319, 2022) tohighlight the potential of machine learning in accelerating reacting flowsimulation. A thorough code validation is conducted via a broad range ofcanonical cases to assess its accuracy and efficiency. The results demonstratethat the convection-diffusion-reaction algorithms implemented in DeepFlame arerobust and accurate for both steady-state and transient processes. In addition,a number of methods aiming to further improve the computational efficiency,e.g. dynamic load balancing and adaptive mesh refinement, are explored. Theirperformances are also evaluated and reported. With the deep learning methodimplemented in this work, a speed-up of two orders of magnitude is achieved ina simple hydrogen ignition case when performed on a medium-end graphicsprocessing unit (GPU). Further gain in computational efficiency is expected forhydrocarbon and other complex fuels. A similar level of acceleration isobtained on an AI-specific chip - deep computing unit (DCU), highlighting thepotential of DeepFlame in leveraging the next-generation computing architectureand hardware.

DeepFlame: A deep learning empowered open-source platform for reacting flow simulations