Side Gust Effects on the Performance of Supersonic Inlet with and without Bleed System Using RANS and URANS

The work outlined in this thesis deals with several topics related to the design and optimization of an axisymmetric bi-conical external compression supersonic air inlet with and without a bleed system at Mach 1.8. It then investigates numerically the effects of side gust on the external and internal flow of the inlet, and its performance using the Reynolds Averaged Navier-Stokes Simulation (RANS) turbulence model. In addition, the response of shock system and flow field inside and outside the inlet to the transient side gust is investigated using the unsteady RANS (URANS) turbulence model. Some preliminary results using a large-eddy simulation (LES) turbulence model, and a comparison of RANS and LES results are also provided. First, a detailed procedure used for the complete design of the supersonic part of the inlet including the cowl angles is outlined. A combination of the Taylor–Maccoll method, characteristic method and existing procedures have been used to design and optimize the geometry of supersonic inlet with and without bleed system. The computations were performed by using a code written in Q-Basic language. The transient and subsonic zones have been designed using the methodology from the existing literature. For the inlet with bleed system, the second cone angle has been further optimized by using the method of characteristics in order to get the oblique shocks’ intersection location very close to, but not at, the cowl lip to have a stable shock system with a relatively small flow spilling over the cowl. Also, for the inlet with bleed system, the bleed system is located at the throat inner wall, and its shape is a combination of a ram scoop and a flush slot in order to provide improvement in total pressure recovery at design and off-design Mach numbers. Next, the effects of side gust on the performance of the inlet without and with a bleed system are investigated by performing 3-D simulations using the ANSYS Fluent 18.2 code and RANS turbulence model. Then, URANS is employed to study the transient side gust effects on some inlet performance parameters and to see how the shock structure and the flow field inside and outside the supersonic inlet respond to such a realistic gust scenario using ANSYS Fluent 19.2 code. A 3D structured mesh with multiple resolutions was generated using the ICEM CFD software. The grid independency tests were performed by using local grid refinements and examining their effect on the inlet performance parameters. In addition, the results from the Fluent code were validated by comparing its predictions with those obtained from the published empirical equations, and those using the commonly used SUPIN and Wind-US codes. Detailed simulation results are then presented to study the complex flow field containing shocks, and characterize the effects of side gust on flow field and inlet performance. The inlet performance is characterized in terms of three parameters, namely, the total pressure recovery (TPR), the inlet mass flow ratio, and flow distortion (FD). There were several important observations. First, the results indicated that simulations capture all the important shock and flow characteristics, both outside and inside the inlet. For the base case (without side gust), the shock waves and flow field are symmetric with respect to the longitudinal axis, and the computed shock system is consistent with the design conditions. Thus, the results show two oblique shocks originating from the compression surfaces and impinging on each other close to the cowl lip, along with a normal shock at the cowl lip, and finally subsonic compression in the diffuser section. Second, the shock wave and flow characteristics are strongly affected by the presence of gust. The major effect of side gust is that the shock structures and the external and internal flows become asymmetric with respect to the longitudinal axis. The oblique shocks in the lower part of the inlet are stronger compared to those in the upper part, and as a consequence, there is a stronger normal shock in the upper part compared to that in the lower part. The asymmetry also leads to a stronger shock-/boundary-layer interaction, causing significant boundary-layer separation, secondary (radial) flow, and a separated flow region in the diffuser section: especially in the lower half. Another effect of side gust is that the normal shock gets located upstream or downstream of the cowl lip, depending on the side gust angle. Consequently, the inlet performance is adversely affected by the gust. The overall effect seems to be stronger at a gust angle of 300 and progressively decreases as the gust angle is increased. For the inlet without bleed system, there is significant reduction in total pressure recovery (more than 20%) and a corresponding increase in the flow distortion for the low side gust angle (α = 300), while the mass flow rate decreases the most for the high side gust angle (α = 900). Third, the results demonstrate the effectiveness of the bleed system in removing the adverse effects of shock waves and their interaction with the boundary layer from the centerbody walls: especially for the side gust cases. This leads to improvement in the inlet performance: especially with respect to flow distortion at the inlet exit, which is noticeably reduced. The results further indicate that the bleed system is not quite as effective in removing the adverse pressure gradient effect for the low gust angle case (α = 300) because the normal shock moves further downstream of the cowl lip and becomes stronger. This causes a significant degradation in TPR (more than 15%) and a corresponding increase in FD for the low gust angle case. This indicates the need for an additional bleed system near the cowl surface in order to remove this adverse effect and further improve inlet performance. From the transient gust study, the results indicate that the flow in the supersonic part (including the oblique shocks part) responds almost instantaneously to the unsteady gust, while the flow in the subsonic part responds slowly and reaches the steady state after 102 ms, which is about 11 times the convective flow time. The bleed plenum also undergo transients due to the presence of transient gust, where the axisymmetric expansion and contraction jet flow in the base case increasingly becomes asymmetric with a thicker jet in the lower part of the plenum and a thinner jet in the upper part as the gust speed ramps up. Another important result is that, the shocks and the flow characteristics, after they reach a steady state, are nearly identical to those predicted by the RANS model used in steady state simulations. For one representative case for the same supersonic inlet with bleed system, simulations are also performed using the LES turbulence model. A preliminary comparison of RANS and LES turbulence models indicates that the RANS model provides fairly reliable predictions of all the mean flow properties, which include the computed flow field including the strengths and locations of oblique and normal shocks, and the inlet performance parameters, such as the TPR, FD, and effective mass flow rate. However, the RANS model unable to predict the fluctuations in the flow properties that LES predicts, but computations with LES model become quite challenging compared to that with RANS model, since the LES model inherently involves unsteady computations, and requires significantly finer mesh than that generally required for RANS.