Andrew T. Corrigan

Towards Efficient, Unsteady, Three-Dimensional Rotating Detonation Engine Simulations

Rotating detonation engines (RDEs) represent an alternative to the extensively studied pulse detonation engines (PDEs) for obtaining propulsion from the high efficiency detonation cycle. Numerical simulations play an important role in understanding the basic physics of the RDE and will be important in optimizing the geometry and flow-field conditions for an RDE, but are necessarily unsteady and three-dimensional. The current paper describes efforts to develop a new code, Propel, for simulating complex engine designs. Propel has support for structured and unstructured meshes, several different numerical algorithms and limiters, and can be run on both CPUs and GPUs in HPC to laptop environments. This paper compares two and three dimensional solutions using Propel for a detonation tube and baseline RDE with our current RDE simulation tool. As an example of the capabilities of the new Propel code, we examine some preliminary calculations of the expansion flow region of a Rotating Detonation Engine as the expansion geometry of the combustion chamber is modified.

Numerical Study of Noise Characteristics in Overexpanded Jet Flows

Abstract : Noise characteristics in shock-containing jets at an overexpanded jet condition have been investigated. Total temperature ratios of 1.0 (cold), 2.0, and 3.0 are considered. The cold jet is a highly screeching jet. Frequency-wavenumber Fourier analysis is employed to examine the wave characteristics of pressure waves along the lip line and also along a near-field conical surface. It is found that the radiating portion of the pressure wave intensity increases with the jet temperature, but the hydrodynamic portion is much less sensitive to the change of the jet temperature. The near-field noise intensity associated with the Mach wave radiation is observed over a large axial distance, and the Mach wave radiation extends to much higher frequencies in heated jets. The peak radiation direction in the cold jet is dominated by the axisymmetric mode, but the directions around the sideline show a much weaker azimuthal dependence. Furthermore, the axial locations of lip-line pressure peak intensities at the screech frequency are near the axial locations of shock-cell tips. A reinforcing loop between upstream/downstream propagating waves and the induced shock-cell coherent oscillatory motion is observed in the highly screeching jet. The formation of this reinforcing loop requires a match of the peak phase velocities between upstream and downstream propagating waves. This phase velocity match exists in the highly screeching cold jet, but not in the weakly screeching heated jets. It appears that the phase velocity match that sustains the reinforcing loop is important to the screech generation, and the phase velocity mismatch in heated jets is believed to be an important cause of the screech intensity reduction.

Area Effects on Rotating Detonation Engine Performance

Prior work shows that the thermodynamic cycle of the rotating detonation engine can be analyzed as a ZND planar detonation that is transformed into a rotating frame of reference. The resulting cycle model is independent of geometry and consistent with the Euler turbomachinery equation. A two-dimensional inviscid numerical model confirms the essential features of the analytical cycle. A subsequent examination confirms the operation of the Euler equation in a three-dimensional numerical simulation and is consistent with the modified ZND model. The geometry of the RDE is varied over nine different geometries of area and radius values. Specific impulse is shown to be proportional to area ratio. The range of exit swirl angles are shown to be proportional to the ratio of exit to inlet radii. The correlation of performance or swirl with geometry suggests other factors are equally important.

A Hybrid Grid Compressible Flow Solver for Large-Scale Supersonic Jet Noise Simulations on Multi-GPU Clusters

ow simulations. Supersonic jet noise simulations require the accurate representation of complex nozzle geometry and thus the use of unstructured grids. While such a grid representation is crucial for the accurate representation of the nozzle geometry, it has the disadvantage of introducing a highly irregular memory access pattern, which violates GPU coalescing requirements resulting in an inecient use of the otherwise high memory bandwidth of GPUs and therefore a bottleneck in computational performance. In order to mitigate this performance bottleneck, a hybrid grid representation is implemented which allows for augmenting the unstructured grid representation in the vicinity of complex nozzle geometry with an ecient structured grid representation in other regions of the ow domain, which is able to fulll GPU coalescing requirements, and thus achieve a signicant improvement in computational performance, leading to a reduction in simulation turnaround time.

Computational Study of the Impact of Chevrons on Noise Characteristics of Imperfectly Expanded Jet Flows

The impact of chevrons on flow field and noise characteristics at an underexpanded jet condition with three jet temperatures has been studied by using large-eddy simulations. It is found that chevrons reduce shock-cell size and shock-cell strength, and also produce additional shock waves with a wavelength almost half the original shock cell size. Increasing the jet temperature reduces the vorticity of induced vortices. In addition, chevrons increase shear-layer thickness, reduce turbulence level in the jet wake region, but have little impact on the jet core length. Double broadband peaks of shock-associated noise are observed in both the nearand far-field sound pressure level distributions. In addition, a large increase of high-frequency components is found near the nozzle exit, and this high-frequency increase is elevated by the heating effect. In the far field, this high-frequency increase is found at both upstream and downstream shallow radiation angles. Chevrons are effective in reducing the broadband shock-associated noise at this underexpanded jet condition, but the effectiveness is reduced as the jet temperature increases, and the least effective region is at the downstream shallow radiation angles, which are upstream of the peak radiation direction. On the other hand, chevrons show a more consistent effectiveness in reducing the peak mixing noise level at those three heating conditions.

Checkpointing Methods for Adjoint-Based Supersonic Jet Noise Reduction

Checkpointing methods have been implemented and optimized to maximize memory usage while minimizing both IO and redudant calculation, when performing adjoint-based optimization using the Jet Engine Noise Reduction (JENRE) code. Unsteady adjoint simulations require the full time history of the large-eddy simulation, in reverse order. Without introducing errors due to compression or interpolation, as has been done in previous work, the present approach allows for much larger adjoint simulations than otherwise possible, providing control or design gradients for only a slightly higher computational cost than performing a large-eddy simulation. The optimal usage of memory space also allows for practical usage of GPU systems with relatively small memory size.

Efficient Supersonic Jet Noise Simulations Using JENRE

This paper provides an overview of the code, JENRE, developed for the computational study of supersonic jet noise reduction under the ONR Jet Noise Reduction project and the NRL Base program on Priority Issues in Naval Air Propulsion. It also presents representative simulations conducted using the code and comparisons of the results of the simulations to experimental data. The JENRE code using the Monotonically Integrated Large Eddy Simulation (MILES) approach is shown to be able to accurately and efficiently simulate the supersonic flows and noise from nozzles representative of military aircraft jets. A key objective of the research effort to reduce the computing time by an order of magnitude has been achieved. The capability of the computations to accurately simulate various potential noise reduction concepts is also demonstrated. These two factors (speed and accuracy) enables the effective use of the JENRE code to investigate potential noise reduction technologies in a cost-effective and timely manner.

Implementation of Thermochemistry and Chemical Kinetics in a GPU-based CFD Code

The implementation of multi-species thermochemistry and chemical kinetics in a GPUbased CFD code is described, focusing on issues which are specific to GPUs. The physical model and its numerical formulation are described in detail. Performance results are presented for two multidimensional test cases: non-reacting supersonic flow over a forward-facing step and the reacting flow of a cellular detonation in a low-pressure H2-O2Ar mixture. The performance results are analyzed to determine the performance when simulations are run on GPUs versus CPUs, the scalability of the solver on both types of compute devices, as well as the cost of the thermochemistry model relative to simpler thermodynamic models.

A Formulation and Implementation of Adjoint-Based Supersonic Jet Noise Reduction

A formulation and implementation of adjoint-based supersonic jet noise reduction is presented, which allows for efficiently computing the gradient of a noise cost function with respect to a given control or design. The cost function is chosen to penalize nearfield pressure fluctuations, the sensitivity of which is then traced back to upstream flow features. This approach, which has been implemented in the Jet Engine Noise Reduction (JENRE) code, is employed to perform unsteady, adjoint, large-eddy simulations involving three-dimensional, high-resolution, unstructured grids.

Fundamental Jet Noise Reduction Science and Technology

This paper provides an overview of the code development and computational study of supersonic jet noise reduction conducted under the ONR Jet Noise Reduction project. Monotonically Integrated Large Eddy Simulations (MILES) is shown to be able to accurately and efficiently simulate the supersonic flows and noise from complex nozzles. A key objective of the research effort to reduce the computing time by an order of magnitude has been achieved. The capability of the computations to accurately simulate various potential noise reduction concepts is also demonstrated. These two factors (speed and accuracy) enables the effective use of the JENRE code to investigate potential noise reduction technologies in a cost-effective and timely manner.