Complex manifold-based turbulent combustion models using in-situ adaptive neural network coupled binary trees

Manifold-based combustion models can decrease the cost of turbulent combustion simulations by projecting the thermochemical state onto a lower-dimensional manifold, allowing the thermochemical state to be computed separately from the flow solver. Solutions for the manifold equations have traditionally been precomputed and pretabulated, resulting in large memory requirements and significant precomputation cost, even for simple problems. In-Situ Adaptive Manifolds (ISAM) enables solutions to the manifold equations to be computed as the simulation progresses and stored using binary trees with In-Situ Adaptive Tabulation (ISAT), allowing for the use of more general models. While ISAT helps reduce the memory requirements compared to pretabulation approaches, as the model complexity grows, the memory requirements of ISAT databases will still eventually become too large. In this work, binary trees within ISAT are pruned and replaced with neural networks to reduce the memory requirements of the ISAT database. Binary tree pruning and neural network training are conducted on-the-fly in LES in the same spirit as the ISAM approach. Memory use, computational timing, and accuracy with the neural network coupled binary trees are compared to the original ISAM implementation using laboratory-scale flames of varying complexity.