SECS-C: Shocklet Ensemble Chemistry Solver for Combustion, in Parallel Hardware Architectures

Existing subsonic combustion models treat the turbulent flame front as a diffusion phenomenon — a mathematical convenience inherited from 1970s hardware constraints rather than a physical necessity. This work proposes SECS-C: Shocklet Ensemble Chemistry Solver for Combustion, a conceptual framework that treats the subsonic flame front as a statistical ensemble of microscopic, localised pressure discontinuities — shocklets — solved via explicit Riemann flux calculations on parallel GPU threads. The approach draws direct analogy to the laminar flamelet concept of Peters (1984), in which a turbulent flame is treated as an ensemble of laminar flamelets, extending the ensemble philosophy one level deeper to the pressure discontinuity structure of the flame front itself. Since the shocklet ensemble is driven by the local pressure gradient magnitude |∇p|, computed at zero additional cost by every Navier-Stokes solver at every cell at every timestep, the framework requires no additional scalar transport equations beyond the Navier-Stokes system. The statistical blending function governing the ensemble is physically motivated by the Rankine-Hugoniot jump conditions rather than chosen for mathematical convenience. SECS-C is the fourth framework emerging from the failure of existing models on a 15 kW CH₄/O₂ non-premixed swirl flame, defended as a Master's thesis at TU Bergakademie Freiberg on 5 June 2026. The preceding frameworks address the same triply constrained problem from different directions: a modified SST-SAS turbulence model addressing the SAS bootstrap deadlock [1], a hyperbolic wave chemistry framework replacing the parabolic diffusion assumption [2], and a thermochemical ray tracing framework recasting chemistry as a GPU ray traversal problem [3]. A fourth prior work proposes LBM-guided domain decomposition as a computational acceleration strategy [4]. All four share the same philosophy: let the physics decide. This is an unvalidated conceptual preprint presented as a direction for future investigation. [1] https://doi.org/10.5281/zenodo.20555825 [2] https://doi.org/10.5281/zenodo.20573169 [3] https://doi.org/10.5281/zenodo.19975933 [4] https://doi.org/10.5281/zenodo.17771873