Computational modeling of dynamic injector response in a Rotating Detonation Engine (RDE)

Rotating Detonation Engines (RDEs) are a form of pressure gain combustion (PGC), offering a promising approach to increase the thermodynamic efficiency of a gas turbine combustor by utilizing a detonation-driven combustion process. In most RDEs, fuel and oxidizer are discretely injected from separate plenums. The discrete fuel/oxidizer injection locations are influenced by the local chamber conditions, leading to mixture inhomogeneity in the combustor. The objective of this study is to develop a dynamic injector response model capable of simulating injector behavior without the need to mesh/resolve the individual injectors. A series of 3D non-reacting computational fluid dynamics (CFD) simulations is used to generate empirical correlations for mass flux and mixture inhomogeneity. These correlations are then implemented as spatially/temporally varying inlet boundary conditions in 2D reacting RDE simulations. The obtained results are compared against experimental data and perfectly premixed simulations for two different RDE geometries, each at two separate operating conditions, focusing on wave speed and static pressure measurements for validation. The injector response model predicted wave speed, which is approximately within 10% of the experimental value. The time-averaged static pressure data determined from the injector response model also lies within the uncertainty limits of experimental measurements, suggesting good agreement between them. The injector response model also provides a computationally cost effective way to incorporate dynamic transient injector response in RDE simulation without meshing/resolving the individual injectors. Additionally, the influence of injector response on wave dynamics, wave structures, and detonation efficiency is investigated. • A dynamic injector response model has been validated against experimental results. The model allows the user to include dynamic fuel/oxidizer injection without meshing/resolving the individual injectors. • The mean detonation velocity for the injector response model is significantly lower than premixed cases and much closer to the measured experimental values. • The EAP for the injector response model shows a significant decrease compared to the premixed condition, thus reducing the available pressure gain. • Dynamic injector response generates a corrugated and curved wave front, and much larger vortical structures are present behind the wave front, reducing post detonation mixing. • Dynamic injector response resulted in a highly distorted and elongated cell size pattern. The local variation in reactant mixture composition causes the triple point to sporadically extinguish and reappear in the domain, leading to this highly distorted wave structure and irregular cell size pattern. • Inhomogeneous mixing due to dynamic injector response results in a lower detonation efficiency, and heat release rates occur over a range of equivalence ratios compared to the perfectly premixed condition.