Design exploration for a single expansion ramp nozzle (SERN) using data mining

The single expansion ramp nozzle is an essential component for hypersonic vehicles to improve their internal/external integral level and produce most of the thrust force. The two-dimensional coupled implicit Reynolds Averaged Navier–Stokes (RANS) equations and the two-equation RNG k–ε turbulent model have been employed to numerically simulate the flow field in a single expansion ramp nozzle, and the interactions between the parametric parameters and the objective functions, namely the thrust force and the lift force, have been investigated by using the data mining technique coupled with a design of the experiment. The obtained results show that the physical model has a good two-dimensional structure, and the numerical results show very good agreement with the available experimental data. At the same time, the grid scale has only a slight impact on the pressure distribution. The influences of the horizontal length of the inner nozzle, the external expansion ramp and the internal cowl expansion on the thrust force performance are substantial, as are the effects of the internal cowl expansion and the external expansion ramp on the lift force performance. Further, optimized configurations of the single expansion ramp nozzle are obtained by using single- and multi-objective design optimization methods coupled with the Kriging surrogate model, and the optimized performances show very good agreement with the numerical predictions. The discrepancies between the optimized performances and the numerical predictions are less than 0.05%, and the method proposed in this paper is efficient in designing and optimizing the nozzle configuration.