Matching Items (7)
Filtering by
- All Subjects: Aerospace Engineering
- Creators: Chen, Kangping
Description
Derived from the necessity to increase testing capabilities of hybrid rocket motor (HRM) propulsion systems for Daedalus Astronautics at Arizona State University, a small-scale motor and test stand were designed and developed to characterize all components of the system. The motor is designed for simple integration and setup, such that both the forward-end enclosure and end cap can be easily removed for rapid integration of components during testing. Each of the components of the motor is removable allowing for a broad range of testing capabilities. While examining injectors and their potential it is thought ideal to obtain the highest regression rates and overall motor performance possible. The oxidizer and fuel are N2O and hydroxyl-terminated polybutadiene (HTPB), respectively, due to previous experience and simplicity. The injector designs, selected for the same reasons, are designed such that they vary only in the swirl angle. This system provides the platform for characterizing the effects of varying said swirl angle on HRM performance.
ContributorsSummers, Matt H (Author) / Lee, Taewoo (Thesis advisor) / Chen, Kangping (Committee member) / Wells, Valana (Committee member) / Arizona State University (Publisher)
Created2013
Description
Multiphase flows are an important part of many natural and technological phe- nomena such as ocean-air coupling (which is important for climate modeling) and the atomization of liquid fuel jets in combustion engines. The unique challenges of multiphase flow often make analytical solutions to the governing equations impos- sible and experimental investigations very difficult. Thus, high-fidelity numerical simulations can play a pivotal role in understanding these systems. This disserta- tion describes numerical methods developed for complex multiphase flows and the simulations performed using these methods. First, the issue of multiphase code verification is addressed. Code verification answers the question "Is this code solving the equations correctly?" The method of manufactured solutions (MMS) is a procedure for generating exact benchmark solutions which can test the most general capabilities of a code. The chief obstacle to applying MMS to multiphase flow lies in the discontinuous nature of the material properties at the interface. An extension of the MMS procedure to multiphase flow is presented, using an adaptive marching tetrahedron style algorithm to compute the source terms near the interface. Guidelines for the use of the MMS to help locate coding mistakes are also detailed. Three multiphase systems are then investigated: (1) the thermocapillary motion of three-dimensional and axisymmetric drops in a confined apparatus, (2) the flow of two immiscible fluids completely filling an enclosed cylinder and driven by the rotation of the bottom endwall, and (3) the atomization of a single drop subjected to a high shear turbulent flow. The systems are simulated numerically by solving the full multiphase Navier- Stokes equations coupled to the various equations of state and a level set interface tracking scheme based on the refined level set grid method. The codes have been parallelized using MPI in order to take advantage of today's very large parallel computational architectures. In the first system, the code's ability to handle surface tension and large tem- perature gradients is established. In the second system, the code's ability to sim- ulate simple interface geometries with strong shear is demonstrated. In the third system, the ability to handle extremely complex geometries and topology changes with strong shear is shown.
ContributorsBrady, Peter, Ph.D (Author) / Herrmann, Marcus (Thesis advisor) / Lopez, Juan (Thesis advisor) / Adrian, Ronald (Committee member) / Calhoun, Ronald (Committee member) / Chen, Kangping (Committee member) / Arizona State University (Publisher)
Created2011
Description
The study of deflagration to detonation transition (DDT) in explosives is of prime importance with regards to insensitive munitions (IM). Critical damage owing to thermal or shock stimuli could translate to significant loss of life and material. The present study models detonation and deflagration of a commonly used granular explosive: cyclotetramethylene-tetranitramine, HMX. A robust literature review is followed by computational modeling of gas gun and DDT tube test data using the Sandia National Lab three-dimensional multi-material Eulerian hydrocode CTH. This dissertation proposes new computational practices and models that aid in predicting shock stimulus IM response. CTH was first used to model experimental data sets of DDT tubes from both Naval Surface Weapons Center and Los Alamos National Laboratory which were initiated by pyrogenic material and a piston, respectively. Analytical verification was performed, where possible, for detonation via empirical based equations at the Chapman Jouguet state with errors below 2.1%, and deflagration via pressure dependent burn rate equations. CTH simulations include inert, history variable reactive burn and Arrhenius models. The results are in excellent agreement with published HMX detonation velocities. Novel additions include accurate simulation of the pyrogenic material BKNO3 and the inclusion of porosity in energetic materials. The treatment of compaction is especially important in modeling precursory hotspots, caused by hydrodynamic collapse of void regions or grain interactions, prior to DDT of granular explosives. The CTH compaction model of HMX was verified within 11% error via a five pronged validation approach using gas gun data and employed use of a newly generated set of P-α parameters for granular HMX in a Mie-Gruneisen Equation of State. Next, the additions of compaction were extended to a volumetric surface burning model of HMX and compare well to a set of empirical burn rates. Lastly, the compendium of detonation and deflagration models was applied to the aforementioned DDT tubes and demonstrate working functionalities of all models, albeit at the expense of significant computational resources. A robust hydrocode methodology is proposed to make use of the deflagration, compaction and detonation models as a means to predict IM response to shock stimulus of granular explosive materials.
ContributorsMahon, Kelly Susan (Author) / Lee, Taewoo (Thesis advisor) / Herrmann, Marcus (Committee member) / Chen, Kangping (Committee member) / Jiao, Yang (Committee member) / Huang, Huei-Ping (Committee member) / Arizona State University (Publisher)
Created2015
Description
This dissertation describes a process for interface capturing via an arbitrary-order, nearly quadrature free, discontinuous Galerkin (DG) scheme for the conservative level set method (Olsson et al., 2005, 2008). The DG numerical method is utilized to solve both advection and reinitialization, and executed on a refined level set grid (Herrmann, 2008) for effective use of processing power. Computation is executed in parallel utilizing both CPU and GPU architectures to make the method feasible at high order. Finally, a sparse data structure is implemented to take full advantage of parallelism on the GPU, where performance relies on well-managed memory operations.
With solution variables projected into a kth order polynomial basis, a k+1 order convergence rate is found for both advection and reinitialization tests using the method of manufactured solutions. Other standard test cases, such as Zalesak's disk and deformation of columns and spheres in periodic vortices are also performed, showing several orders of magnitude improvement over traditional WENO level set methods. These tests also show the impact of reinitialization, which often increases shape and volume errors as a result of level set scalar trapping by normal vectors calculated from the local level set field.
Accelerating advection via GPU hardware is found to provide a 30x speedup factor comparing a 2.0GHz Intel Xeon E5-2620 CPU in serial vs. a Nvidia Tesla K20 GPU, with speedup factors increasing with polynomial degree until shared memory is filled. A similar algorithm is implemented for reinitialization, which relies on heavier use of shared and global memory and as a result fills them more quickly and produces smaller speedups of 18x.
With solution variables projected into a kth order polynomial basis, a k+1 order convergence rate is found for both advection and reinitialization tests using the method of manufactured solutions. Other standard test cases, such as Zalesak's disk and deformation of columns and spheres in periodic vortices are also performed, showing several orders of magnitude improvement over traditional WENO level set methods. These tests also show the impact of reinitialization, which often increases shape and volume errors as a result of level set scalar trapping by normal vectors calculated from the local level set field.
Accelerating advection via GPU hardware is found to provide a 30x speedup factor comparing a 2.0GHz Intel Xeon E5-2620 CPU in serial vs. a Nvidia Tesla K20 GPU, with speedup factors increasing with polynomial degree until shared memory is filled. A similar algorithm is implemented for reinitialization, which relies on heavier use of shared and global memory and as a result fills them more quickly and produces smaller speedups of 18x.
ContributorsJibben, Zechariah J (Author) / Herrmann, Marcus (Thesis advisor) / Squires, Kyle (Committee member) / Adrian, Ronald (Committee member) / Chen, Kangping (Committee member) / Treacy, Michael (Committee member) / Arizona State University (Publisher)
Created2015
Description
Modern gas turbines operate at high mainstream gas temperatures and pressures, which requires high durability materials. A method of preventing these hot gases from leaking into the turbine cavities is essential for improved reliability and cost reduction. Utilizing bleed-off air from the compressor to cool internal components has been a common solution, but at the cost of decreasing turbine performance. The present work thoroughly describes the complex flow field between the mainstream gas and a single rotor-stator disk cavity, and mechanisms of mainstream gas ingestion. A combined approach of experimental measurement and numerical simulation are performed on the flow in a single-stage model gas turbine. Mainstream gas ingestion into the cavity is further reduced by utilizing two axially overlapping seal rings, one on the rotor disk and the other on the stator wall. Secondary purge air is injected into the rotor-stator cavity pre-swirled through the stator radially inboard of the two seal rings. Flow field predictions from the simulations are compared against experimental measurements of static pressure, velocity, and tracer gas concentration acquired in a nearly identical model configuration. Operational conditions were performed with a main airflow Reynolds number of 7.86e4 and a rotor disk speed of 3000rpm. Additionally the rotational Reynolds number was 8.74e5 with a purge air nondimensional flow rate cw=4806. The simulation models a 1/14 rotationally periodic sector of the turbine rig, consisting of four rotor blades and four stator vanes. Gambit was used to generate the three-dimensional unstructured grids ranging from 10 to 20 million cells. Effects of turbulence were modeled using the single-equation Spalart-Allmaras as well as the realizable k-epsilon models. Computations were performed using FLUENT for both a simplified steady-state and subsequent time-dependent formulation. Simulation results show larger scale structures across the entire sector angle inside the cavity and certain unsteady mainstream ingestion mechanisms are realized from the tracer gas. Simulated velocity distributions were scrutinized against Particle Image Velocimetry plots in the rotor-stator cavity and are in reasonable agreement with all of the measurements.
ContributorsDunn, Dennis Martin (Author) / Squires, Kyle D (Thesis advisor) / Roy, Ramendra P (Committee member) / Chen, Kangping (Committee member) / Arizona State University (Publisher)
Created2010
Description
This dissertation introduces FARCOM (Fortran Adaptive Refiner for Cartesian Orthogonal Meshes), a new general library for adaptive mesh refinement (AMR) based on an unstructured hexahedral mesh framework. As a result of the underlying unstructured formulation, the refinement and coarsening operators of the library operate on a single-cell basis and perform in-situ replacement of old mesh elements. This approach allows for h-refinement without the memory and computational expense of calculating masked coarse grid cells, as is done in traditional patch-based AMR approaches, and enables unstructured flow solvers to have access to the automated domain generation capabilities usually only found in tree AMR formulations.
The library is written to let the user determine where to refine and coarsen through custom refinement selector functions for static mesh generation and dynamic mesh refinement, and can handle smooth fields (such as level sets) or localized markers (e.g. density gradients). The library was parallelized with the use of the Zoltan graph-partitioning library, which provides interfaces to both a graph partitioner (PT-Scotch) and a partitioner based on Hilbert space-filling curves. The partitioned adjacency graph, mesh data, and solution variable data is then packed and distributed across all MPI ranks in the simulation, which then regenerate the mesh, generate domain decomposition ghost cells, and create communication caches.
Scalability runs were performed using a Leveque wave propagation scheme for solving the Euler equations. The results of simulations on up to 1536 cores indicate that the parallel performance is highly dependent on the graph partitioner being used, and differences between the partitioners were analyzed. FARCOM is found to have better performance if each MPI rank has more than 60,000 cells.
The library is written to let the user determine where to refine and coarsen through custom refinement selector functions for static mesh generation and dynamic mesh refinement, and can handle smooth fields (such as level sets) or localized markers (e.g. density gradients). The library was parallelized with the use of the Zoltan graph-partitioning library, which provides interfaces to both a graph partitioner (PT-Scotch) and a partitioner based on Hilbert space-filling curves. The partitioned adjacency graph, mesh data, and solution variable data is then packed and distributed across all MPI ranks in the simulation, which then regenerate the mesh, generate domain decomposition ghost cells, and create communication caches.
Scalability runs were performed using a Leveque wave propagation scheme for solving the Euler equations. The results of simulations on up to 1536 cores indicate that the parallel performance is highly dependent on the graph partitioner being used, and differences between the partitioners were analyzed. FARCOM is found to have better performance if each MPI rank has more than 60,000 cells.
ContributorsBallesteros, Carlos Alberto (Author) / Herrmann, Marcus (Thesis advisor) / Adrian, Ronald (Committee member) / Chen, Kangping (Committee member) / Huang, Huei-Ping (Committee member) / Lopez, Juan (Committee member) / Arizona State University (Publisher)
Created2019
Description
Drainage flow of a viscous compressible gas from a semi-sealed narrow conduit is a pore-scale model for studying the fundamental flow physics of fluid recovery from a porous reservoir without using fluid injection. Thermal effect has been routinely neglected for these flows in the traditional petroleum engineering literature. Since the motion is entirely driven by volumetric expansion, temperature change always accompanies the density change. This thesis examines such thermal effects on the drainage flow.
Thermal drainage flow is first studied by simultaneously solving the linearized continuity, momentum and energy equations for adiabatic walls. It is shown that even in the absence of an imposed temperature drop, gas expansion induces a transient temperature decrease inside the channel, which slows down the drainage process compared to the isothermal model and Lighthill’s model. For a given density drop, gas drains out faster as the initial-to-final temperature ratio increases; and the transient density can undershoot the final equilibrium value. A parametric study is then carried out to explore the influence of various thermal boundary conditions on drainage flow. It is found that as the wall transitions from adiabatic to isothermal condition, the excess density changes from a plane wave solution to a non-plane wave solution and the drainage rate increases. It is shown that when the exit is also cooled and the wall is non-adiabatic, the total recovered fluid mass exceeds the amount based on the isothermal theory which is determined by the initial and final density difference alone. Finally, a full numerical simulation is conducted to mimic the channel-reservoir system using the finite volume method. The Ghost-Cell Navier-Stokes Characteristic Boundary Condition technique is applied at the far end of the truncated reservoir, which is an open boundary. The results confirm the conclusions of the linear theory.
Thermal drainage flow is first studied by simultaneously solving the linearized continuity, momentum and energy equations for adiabatic walls. It is shown that even in the absence of an imposed temperature drop, gas expansion induces a transient temperature decrease inside the channel, which slows down the drainage process compared to the isothermal model and Lighthill’s model. For a given density drop, gas drains out faster as the initial-to-final temperature ratio increases; and the transient density can undershoot the final equilibrium value. A parametric study is then carried out to explore the influence of various thermal boundary conditions on drainage flow. It is found that as the wall transitions from adiabatic to isothermal condition, the excess density changes from a plane wave solution to a non-plane wave solution and the drainage rate increases. It is shown that when the exit is also cooled and the wall is non-adiabatic, the total recovered fluid mass exceeds the amount based on the isothermal theory which is determined by the initial and final density difference alone. Finally, a full numerical simulation is conducted to mimic the channel-reservoir system using the finite volume method. The Ghost-Cell Navier-Stokes Characteristic Boundary Condition technique is applied at the far end of the truncated reservoir, which is an open boundary. The results confirm the conclusions of the linear theory.
ContributorsHuang, Wei (Author) / Chen, Kangping (Thesis advisor) / Huang, Huei-Ping (Committee member) / Herrmann, Marcus (Committee member) / Calhoun, Ronald (Committee member) / Baer, Steven (Committee member) / Arizona State University (Publisher)
Created2020