During unstart, the rapid upstream propagation of a hypersonic engine's inlet shock system can be clearly seen through inlet pressure measurements. Specifically, the magnitude of the pressure readings suddenly and dramatically increases as soon as the leading edge of the shock system passes the measurement location. A change detection algorithm can monitor the pressure time history at a given sensing location and determine when an abrupt pressure rise occurs. If this kind of information can be obtained at various sensing locations distributed throughout the inlet then a feedback control scheme has an improved basis upon which to make actuation decisions for preventing unstart. In this thesis a variety of change detection algorithms have been implemented and tested on multiple sources of experimental high-speed pressure transducer data. The performance of these algorithms is compared and suitability of each algorithm for the general unstart problem is discussed. Attempts to model the transient dynamics governing the unstart process have also been made through the use of system identification techniques. The result of these system identification efforts is a partially nonlinear mathematical model that describes shock motion through pressure signals. The process reveals that the nonlinear behavior can be separated from the linear with relative ease. Related attempts are then made to create a model where the nonlinear portion has been specified leaving only the linear portion to be determined by system identification. The modeling and identification process specific to the unstart data used is discussed and successful models are presented for both cases. / text

No description available.

Hypersonic inlet unstart remains a major technical obstacle in the successful implementation of hypersonic air-breathing propulsion systems such as ramjets and scramjets. Unstart occurs when combustor-induced pressure fluctuations lead to rapid expulsion of the shock system from the isolator, and is associated with loss of thrust. The research presented here attempts to mitigate this behavior through the design and implementation of a closed-loop control scheme that regulates shock location within a Mach 1.8 wind tunnel isolator test section. To localize the position of the shock within the isolator, a set of high frequency Kulite pressure transducers are used to measure the static pressure at various points along the wind tunnel test section. A novel Kalman filter based approach is utilized, which fuses the estimates from two distinct shock localization algorithms running at 250 Hz to determine the location of the shock in real time. The primary shock localization algorithm is a geometrical shock detection scheme that can estimate the position of the shock system even when it is located between pressure transducers. The second algorithm utilizes a sum-of-pressures technique that can be calibrated by the geometrical algorithm in real time. The closed-loop controller generates commands every 100 ms to actuate a motorized flap downstream of the test section in an effort to regulate the shock to the desired location. The closed-loop control implementation utilized a simple logic-based controller as well as a Proportional-Integral (PI) and a Proportional-Derivative (PD) Controller. In addition to the implementation of control algorithms, the importance of various design criteria necessary to achieve satisfactory control performance is explored including parameters such as pressure transducer spacing, shock localization speed, flap-motor actuation speed and actuator resolution. Experimental results are presented for various test scenarios such as regulation of the shock location in the presence of stagnation pressure disturbances as well as tracking of time-varying step inputs. Performance and robustness properties of the tested control implementations are discussed. Further areas of improvement for the closed-loop control system in both hardware and software are discussed, and the need for reduced-order dynamics-based controllers is presented. / text

In the pursuit of developing hypersonic cruise vehicles, unstart is a major roadblock to achieving stable flight. Unstart occurs when a sudden instability in the combustor of a vehicle’s propulsion system creates an instantaneous pressure rise that initiates a shock. This shock travels upstream out of the inlet of the vehicle, until it is ejected from the inlet and creates a standing shockwave that chokes the flow entering the vehicle, thereby greatly reducing its propulsive capability. In severe cases, this can lead to the loss of the vehicle. This thesis presents the results of a computational study of the dynamics of unstart near Mach 5 and presents some possible precursor signals that may indicate its presence in flight. Using SU2, an open-source CFD code developed at Stanford University, the Unsteady Reynolds-Averaged Navier-Stokes equations are used to develop a model for flow in a scramjet inlet-isolator geometry, both in the fully started state and during unstart. The results of these calculations were compared against experimental data collected by J. Wagner, at the University of Texas, Austin. In the present computations, unstart was initiated through the use of an artificial body force, which mimicked a moveable flap used in the experiments. Once the results of the code were validated against these experiments, a selection of parametric studies were conducted to determine how the design of the inlet-isolator by Wagner affected the flow, and thus how generalizable the results can be. In addition, precursor signals indicative of unstart were identified for further study and examined in the different parametric studies. It was found that a thick boundary layer is conducive to a stronger precursor signal and a slower unstart. In addition, an aspect ratio closer to 1:1 promotes flow mixing and reduces the unstart speed and strength. Moreover, an aspect ratio in this range reduces the precursor signal strength but, if a thick boundary layer is present, will smear the signal out over a larger area, potentially making it easier to detect. <br>

The dynamics of the unstart process in inlet / isolator models mounted to the floor of a Mach 5 wind tunnel are investigated experimentally. The most extensively studied model has an inlet section that contains a 6-degree compression ramp and the isolator is a rectangular straight duct that is 25.4 mm high by 50.8 mm wide by 242.3 mm long. Unstart is initiated by raising a motor-driven flap that is located at the downstream end of the isolator section. Unstart proceeds with the formation of a shock system that propagates upstream at an average velocity of about 37 m/s (in the lab frame of reference), which is five percent of the freestream velocity. Unstart is seen to be associated with strong shock-induced separation that leads to reverse flow velocities up to about 300 m/s as measured by PIV. Both the schlieren imaging and PIV data suggest the dynamics and flow structure of the unstart process are dependent on inlet geometry. Furthermore, the PIV data indicate the unstart process to be highly three-dimensional. Finally, tripping the ceiling and sidewall boundary layers was seen to result in slower unstart processes. In addition, results are presented for 0-degree (no inlet) and 8-degree inlet / isolator models. In the 0-degree model, the experimental data show that the flow structure and propagation velocities of the unstart shock system are much more constant than those measured in unstart events with an inlet. In addition, an increased inlet compression angle appears to result in an increased unstart propagation velocity in the isolator. This is possibly related to the fact that with an increased compression ramp angle, the unstart shock system propagates against a lower momentum opposing flow. Furthermore, the inlet geometry is also seen to affect the flow that follows the unstart process. Experiments were also conducted with each of the three inlets attached to a shortened isolator. The short-isolator experiments showed it was possible to form a stable high-compression shock system in the isolator by raising the flap. This was not the case in longer isolator tests. / text

ABSTRACT CONCEPTUAL INTERNAL DESIGN AND COMPUTATIONAL FLUID DYNAMICS ANALYSIS OF A SUPERSONIC INLET ALEMDAROgLU, Mine M. S., Department of Aerospace Engineering Supervisor: Prof. Dr. Yusuf &Ouml / ZY&Ouml / R&Uuml / K May 2005, 144 pages In this thesis, the conceptual internal design of the air inlet of a supersonic, high altitude, solid propellant ramjet cruise missile is performed. Inviscid, compressible CFD analysis of the designed inlet is made in order to obtain qualitative and quantitative performance characteristics of the inlet at different operating conditions. The conceptual design of the inlet is realized by using analytical relations and equations, correlations derived from numerous available past experimental data and state-of-the-art design examples. The performance estimation of the designed inlet at different operating conditions is done by using one and two dimensional gas dynamics equations. The results of the performance estimation study are compared with the results of the CFD analysis and these results are discussed in detail. A commercial tool, CFD-FASTRAN&Ograve / , is used for the CFD analysis. Inlet flow phenomena such as, different shock patterns and shock positions, performance degradation at off-design operating conditions and inlet unstart are observed. Keywords: Supersonic Inlet, Ramjet, CFD, Inlet Performance Characteristics, Operating Conditions, Unstart

No description available.

The main objective of this dissertation is to develop large-eddy simulation (LES) based computational tools for supersonic inlet and combustor design. In the recent past, LES methodology has emerged as a viable tool for modeling turbulent combustion. LES computes the large scale mixing process accurately, thereby providing a better starting point for small-scale models that describe the combustion process. In fact, combustion models developed in the context of Reynolds-averaged Navier Stokes (RANS) equations exhibit better predictive capability when used in the LES framework. The development of a predictive computational tool based on LES will provide a significant boost to the design of scramjet engines. Although LES has been used widely in the simulation of subsonic turbulent flows, its application to high-speed flows has been hampered by a variety of modeling and numerical issues. In this work, we develop a comprehensive LES methodology for supersonic flows, focusing on the simulation of scramjet engine components. This work is divided into three sections. First, a robust compressible flow solver for a generalized high-speed flow configuration is developed. By using carefully designed numerical schemes, dissipative errors associated with discretization methods for high-speed flows are minimized. Multiblock and immersed boundary method are used to handle scramjet-specific geometries. Second, a new combustion model for compressible reactive flows is developed. Subsonic combustion models are not directly applicable in high-speed flows due to the coupling between the energy and velocity fields. Here, a probability density function (PDF) approach is developed for high-speed combustion. This method requires solution to a high dimensional PDF transport equation, which is achieved through a novel direct quadrature method of moments (DQMOM). The combustion model is validated using experiments on supersonic reacting flows. Finally, the LES methodology is used to study the inlet-isolator component of a dual-mode scramjet. The isolator is a critical component that maintains the compression shock structures required for stable combustor operation in ramjet mode. We simulate unsteady dynamics inside an experimental isolator, including the propagation of an unstart event that leads to loss of compression. Using a suite of simulations, the sensitivity of the results to LES models and numerical implementation is studied. / text

No description available.

No description available.