Abstract

Abstract Current gas turbine technology utilizes deflagration combustion, but an rotating detonation engine (RDE), if integrated successfully, has the potential to significantly increase thermal efficiency. The shock-laden RDE exit flow is highly unsteady and spatially non-uniform (hydrodynamically and thermally), with a high degree of periodicity. In contrast, gas turbines are designed to operate with relatively small velocity and temperature fluctuations at the inlet. The objective of the current work is to optimize and study the impact of area profiling in an annular RDE configuration. A computational approach is used to condition the flow within the RDE channel by strategically constricting the flow area to convert the unsteady supersonic/subsonic flow exiting the combustor into a relatively uniform steady flow with minimum pressure loss. In the current work, a 3D non-reacting optimization methodology is developed for the wall profiling of the combustor. The emphasis is to reduce the computational cost needed for design optimization. A Design of Experiment (DoE) study is conducted, and the geometry is parametrized for the constriction on the inner wall. It consists of a constant area section up to the detonation wave height, followed by a parabolic area profile (rapid reduction in area), and finally linear profile (gradual reduction in area) till the RDE exit. Three geometric parameters are varied, which include the length of the parabolic section, the area ratio at the end of the parabolic section (with respect to the constant area section), and the throat area ratio (with respect to the constant area section) at the exit. The output parameters include three primary indicators: equivalent available pressure (EAP), axial velocity unsteadiness, and circumferential velocity unsteadiness. The initial points in the DoE are coarsely populated, and an initial response surface is generated using the Kriging method. Next, the response surface is refined based on the adaptive sampling technique. Finally, design optimization is performed to determine the optimal design configuration that maximizes the pressure gain and minimizes the unsteadiness. A CFD simulation with constant area cross-section for pure CH4 with enriched-oxygen air is validated with experimental data followed by the validation of the rapid to gradual area reduction (RTG) profile simulation with experiments. The benefits of the RTG profile are compared with a constant cross-section area to delineate the impact of area reduction on flow uniformity at the exit of the RDE.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call