SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
The applications of computers and data processing to astronomy are discussed. Among the topics covered are the emerging national information infrastructure, workstations and supercomputers, supertelescopes, digital astronomy, astrophysics in a numerical laboratory, community software, archiving of ground-based observations, dynamical simulations of complex systems, plasma astrophysics, and the remote control of fourth dimension supercomputers.
The structure and evolution of self-excited subsonic periodic arrays of jets of constant and variable density are studied using spectral-element direct numerical simulations. The governing equation of motion is presented, and a method based on spectral element discretizations appropriate for simulating arbitrarily complex geometry jets and large density variations for subsonic flows is developed. Variable density fields are found to be more unstable than the corresponding uniform density fields with much higher rms values; as a result, their spreading is also considerably larger. There is a dramatic increase in spreading after a few pairings occur. Findings presented for low and high side-momentum flux reveal a shifting of the origin of instability from the near-field to the far-field, respectively, and suggest possible routes of stabilization.
The lack of available wideband digital links as well as the complexity of implementation of bandwidth efficient digital video CODECs (encoder/decoder) has worked to keep the cost of digital television transmission too high to compete with analog methods. Terrestrial and satellite video service providers, however, are now recognizing the potential gains that digital video compression offers and are proposing to incorporate compression systems to increase the number of available program channels. NASA is similarly recognizing the benefits of and trend toward digital video compression techniques for transmission of high quality video from space and therefore, has developed a digital television bandwidth compression algorithm to process standard National Television Systems Committee (NTSC) composite color television signals. The algorithm is based on differential pulse code modulation (DPCM), but additionally utilizes a non-adaptive predictor, non-uniform quantizer and multilevel Huffman coder to reduce the data rate substantially below that achievable with straight DPCM. The non-adaptive predictor and multilevel Huffman coder combine to set this technique apart from other DPCM encoding algorithms. All processing is done on a intra-field basis to prevent motion degradation and minimize hardware complexity. Computer simulations have shown the algorithm will produce broadcast quality reconstructed video at an average transmission rate of 1.8 bits/pixel. Hardware implementation of the DPCM circuit, non-adaptive predictor and non-uniform quantizer has been completed, providing realtime demonstration of the image quality at full video rates. Video sampling/reconstruction circuits have also been constructed to accomplish the analog video processing necessary for the real-time demonstration. Performance results for the completed hardware compare favorably with simulation results. Hardware implementation of the multilevel Huffman encoder/decoder is currently under development along with implementation of a buffer control algorithm to accommodate the variable data rate output of the multilevel Huffman encoder. A video CODEC of this type could be used to compress NTSC color television signals where high quality reconstruction is desirable (e.g., Space Station video transmission, transmission direct-to-the-home via direct broadcast satellite systems or cable television distribution to system headends and direct-to-the-home).
The use of real-time simulation at the NASA facility is reviewed specifically with regard to hardware, software, and the use of a fiberoptic-based digital simulation network. The network hardware includes supercomputers that support 32- and 64-bit scalar, vector, and parallel processing technologies. The software include drivers, real-time supervisors, and routines for site-configuration management and scheduling. Performance specifications include: (1) benchmark solution at 165 sec for a single CPU; (2) a transfer rate of 24 million bits/s; and (3) time-critical system responsiveness of less than 35 msec. Simulation applications include the Differential Maneuvering Simulator, Transport Systems Research Vehicle simulations, and the Visual Motion Simulator. NASA is shown to be in the final stages of developing a high-performance computing system for the real-time simulation of complex high-performance aircraft.
The paper reviews the factors that affect the versatility, computational efficiency, and accuracy of the various direct simulation methods that are being applied to problems in rarefied gas dynamics. The influence of computer hardware developments on the cost-effectiveness is discussed, as are the software factors that influence the speed with which simulation programs can be applied to new problems. Particular attention is given to a novel method that improves the efficiency of the simulation for complex 3D problems. The topical issues of reduced resolution procedures, stored lists, and variance reduction schemes are discussed in some detail.
We have completed computer simulations that model shock effects in the venusian atmosphere caused during the passage of two cometlike bodies 100 m and 1000 m in diameter and an asteroidlike body 10 km in diameter. Our objective is to examine hypervelocity-generated shock effects in the venusian atmosphere for bodies of different types and sizes in order to understand the following: (1) their deceleration and depth of penetration through the atmosphere; and (2) the onset of possible ground-surface shock effects such as splotches, craters, and ejecta formations. The three bodies were chosen to include both a range of general conditions applicable to Venus as well as three specific cases of current interest. These calculations use a new multiphase computer code (DICE-MAZ) designed by California Research & Technology for shock-dynamics simulations in complex environments. The code was tested and calibrated in large-scale explosion, cratering, and ejecta research. It treats a wide range of different multiphase conditions, including material types (vapor, melt, solid), particle-size distributions, and shock-induced dynamic changes in velocities, pressures, temperatures (internal energies), densities, and other related parameters, all of which were recorded in our calculations.
Many operational limitations of helicopters and other rotary-wing aircraft are due to nonlinear aerodynamic phenomena incuding unsteady, three-dimensional transonic and separated flow near the surfaces and highly vortical flow in the wakes of rotating blades. Modern computational fluid dynamics (CFD) technology offers new tools to study and simulate these complex flows. However, existing Euler and Navier-Stokes codes have to be modified significantly for rotorcraft applications, and the enormous computational requirements presently limit their use in routine design applications. Nevertheless, the Euler/Navier-Stokes technology is progressing in anticipation of future supercomputers that will enable meaningful calculations to be made for complete rotorcraft configurations.
This paper investigates the accuracy with which the mass of a small body can be determined from a flyby. Due to the nearly linear nature of hyperbolic flybys of small bodies, analytic solutions can be derived. Four analytic solutions are developed, each of increasing complexity with different applications. These solutions are compared to a complex computer simulation for accuracy. When these methods are applied to nearly linear trajectories, the analytic solutions agree with the computer simulation results to within 4.0 percent except in one boundary case which is discussed in the text. Also described is the linear range for which these equations are applicable. These solutions are then applied to the proposed asteroid flyby of the CRAF mission to predict the mass estimate accuracy.
Over recent years, Adaptive Mesh Refinement (AMR) algorithms which dynamically match the local resolution of the computational grid to the numerical solution being sought have emerged as powerful tools for solving problems that contain disparate length and time scales. In particular, several workers have demonstrated the effectiveness of employing an adaptive, block-structured hierarchical grid system for simulations of complex shock wave phenomena. Unfortunately, from the parallel algorithm developer's viewpoint, this class of scheme is quite involved; these schemes cannot be distilled down to a small kernel upon which various parallelizing strategies may be tested. However, because of their block-structured nature such schemes are inherently parallel, so all is not lost. In this paper we describe the method by which Quirk's AMR algorithm has been parallelized. This method is built upon just a few simple message passing routines and so it may be implemented across a broad class of MIMD machines. Moreover, the method of parallelization is such that the original serial code is left virtually intact, and so we are left with just a single product to support. The importance of this fact should not be underestimated given the size and complexity of the original algorithm.
The paper identifies speed, agility, human interface, generation of sensitivity information, task decomposition, and data transmission (including storage) as important attributes for a computer environment to have in order to support engineering design effectively. It is argued that when examined in terms of these attributes the presently available environment can be shown to be inadequate a radical improvement is needed, and it may be achieved by combining new methods that have recently emerged from multidisciplinary design optimization (MDO) with massively parallel processing computer technology. The caveat is that, for successful use of that technology in engineering computing, new paradigms for computing will have to be developed - specifically, innovative algorithms that are intrinsically parallel so that their performance scales up linearly with the number of processors. It may be speculated that the idea of simulating a complex behavior by interaction of a large number of very simple models may be an inspiration for the above algorithms, the cellular automata are an example. Because of the long lead time needed to develop and mature new paradigms, development should be now, even though the widespread availability of massively parallel processing is still a few years away.
The relative roles of large-scale scientific computers and physical experiments in several science and engineering disciplines are discussed. Increasing dependence on computers is shown to be motivated both by the rapid growth in computer speed and memory, which permits accurate numerical simulation of complex physical phenomena, and by the rapid reduction in the cost of performing a calculation, which makes computation an increasingly, attractive complement to experimentation. Computer speed and memory requirements are presented for selected areas of such disciplines as fluid dynamics, aerodynamics, aerothermodynamics, chemistry, atmospheric sciences, astronomy, and astrophysics, together with some examples of the complementary nature of computation and experiment. Finally the impact of the emerging role of computers in the technical disciplines is discussed in terms of both the requirements for experimentation and the attainment of previously inaccessible information on physical processes.
The paper identifies speed, agility, human interface, generation of sensitivity information, task decomposition, and data transmission (including storage) as important attributes for a computer environment to have in order to support engineering design effectively. It is argued that when examined in terms of these attributes the presently available environment can be shown to be inadequate. A radical improvement is needed, and it may be achieved by combining new methods that have recently emerged from multidisciplinary design optimisation (MDO) with massively parallel processing computer technology. The caveat is that, for successful use of that technology in engineering computing, new paradigms for computing will have to be developed - specifically, innovative algorithms that are intrinsically parallel so that their performance scales up linearly with the number of processors. It may be speculated that the idea of simulating a complex behaviour by interaction of a large number of very simple models may be an inspiration for the above algorithms; the cellular automata are an example. Because of the long lead time needed to develop and mature new paradigms, development should begin now, even though the widespread availability of massively parallel processing is still a few years away.
The paper identifies speed, agility, human interface, generation of sensitivity information, task decomposition, and data transmission (including storage) as important attributes for a computer environment to have in order to support engineering design effectively. It is argued that when examined in terms of these attributes the presently available environment can be shown to be inadequate. A radical improvement is needed, and it may be achieved by combining new methods that have recently emerged from multidisciplinary design optimisation (MDO) with massively parallel processing computer technology. The caveat is that, for successful use of that technology in engineering computing, new paradigms for computing will have to be developed - specifically, innovative algorithms that are intrinsically parallel so that their performance scales up linearly with the number of processors. It may be speculated that the idea of simulating a complex behaviour by interaction of a large number of very simple models may be an inspiration for the above algorithms; the cellular automata are an example. Because of the long lead time needed to develop and mature new paradigms, development should begin now, even though the widespread availability of massively parallel processing is still a few years away.
The ultimate goal of virtual acoustics is to simulate the complex acoustic field experienced by a listener freely moving around within an environment. This paper discusses some of the engineering constraints that may be faced during implementation and the perceptual consequences of these constraints. In particular, the perceptual impact of parameters like the update rate and overall system latency of interactive spatial audio systems is addressed.
Immersive, three-dimensional displays are increasingly becoming a goal of advanced human-machine interfaces. While the technology for achieving truly useful multisensory environments is still being developed, techniques for generating three-dimensional sound are now both sophisticated and practical enough to be applied to acoustic displays. The ultimate goal of virtual acoustics is to simulate the complex acoustic field experienced by a listener freely moving around within an environment. Of course, such complexity, freedom of movement and interactivity is not always possible in a 'true' virtual environment, much less in lower-fidelity multimedia systems. However, many of the perceptual and engineering constraints (and frustrations) that researchers, engineers and listeners have experienced in virtual audio are relevant to multimedia. In fact, some of the problems that have been studied will be even more of an issue for lower fidelity systems that are attempting to address the requirements of a huge, diverse and ultimately unknown audience. Examples include individual differences in head-related transfer functions, A lack of real interactively (head-tracking) in many multimedia displays, and perceptual degradation due to low sampling rates and/or low-bit compression. This paper discusses some of the engineering constraints faced during implementation of virtual acoustic environments and the perceptual consequences of these constraints. Specific examples are given for NASA applications such as telerobotic control, aeronautical displays, and shuttle launch communications. An attempt will also be made to relate these issues to low-fidelity implementations such as the internet.
Immersive, three-dimensional displays are increasingly becoming a goal of advanced human-machine interfaces. While the technology for achieving truly useful multisensory environments is still being developed, techniques for generating three-dimensional sound are now both sophisticated and practical enough to be applied to acoustic displays. The ultimate goal of virtual acoustics is to simulate the complex acoustic field experienced by a listener freely moving around within an environment. Of course, such complexity, freedom of movement and interactively is not always possible in a "true" virtual environment, much less in lower-fidelity multimedia systems. However, many of the perceptual and engineering constraints (and frustrations) that researchers, engineers and listeners have experienced in virtual audio are relevant to multimedia. In fact, some of the problems that have been studied will be even more of an issue for lower fidelity systems that are attempting to address the requirements of a huge, diverse and ultimately unknown audience. Examples include individual differences in head-related transfer functions, a lack of real interactively (head-tracking) in many multimedia displays, and perceptual degradation due to low sampling rates and/or low-bit compression. This paper discusses some of the engineering Constraints faced during implementation of virtual acoustic environments and the perceptual consequences of these constraints. Specific examples are given for NASA applications such as telerobotic control, aeronautical displays, and shuttle launch communications. An attempt will also be made to relate these issues to low-fidelity implementations such as the internet.
The overset grid methodology has significantly reduced time-to-solution of high-fidelity computational fluid dynamics (CFD) simulations about complex aerospace configurations. The solution process resolves the geometrical complexity of the problem domain by using separately generated but overlapping structured discretization grids that periodically exchange information through interpolation. However, high performance computations of such large-scale realistic applications must be handled efficiently on state-of-the-art parallel supercomputers. This paper analyzes the effects of various performance enhancement techniques on the parallel efficiency of an overset grid Navier-Stokes CFD application running on an SGI Origin2000 machine. Specifically, the role of asynchronous communication, grid splitting, and grid grouping strategies are presented and discussed. Results indicate that performance depends critically on the level of latency hiding and the quality of load balancing across the processors.
Accurate numerical simulations of complex multiscale compressible viscous flows, especially high speed turbulence combustion and acoustics, demand high order schemes with adaptive numerical dissipation controls. Standard high resolution shock-capturing methods are too dissipative to capture the small scales and/or long-time wave propagations without extreme grid refinements and small time steps. An integrated approach for the control of numerical dissipation in high order schemes for the compressible Euler and Navier-Stokes equations has been developed and verified by the authors and collaborators. These schemes are suitable for the problems in question. Basically, the scheme consists of sixth-order or higher non-dissipative spatial difference operators as the base scheme. To control the amount of numerical dissipation, multiresolution wavelets are used as sensors to adaptively limit the amount and to aid the selection and/or blending of the appropriate types of numerical dissipation to be used. Magnetohydrodynamics (MHD) waves play a key role in drag reduction in highly maneuverable high speed combat aircraft, in space weather forecasting, and in the understanding of the dynamics of the evolution of our solar system and the main sequence stars. Although there exist a few well-studied second and third-order high-resolution shock-capturing schemes for the MHD in the literature, these schemes are too diffusive and not practical for turbulence/combustion MHD flows. On the other hand, extension of higher than third-order high-resolution schemes to the MHD system of equations is not straightforward. Unlike the hydrodynamic equations, the inviscid MHD system is non-strictly hyperbolic with non-convex fluxes. The wave structures and shock types are different from their hydrodynamic counterparts. Many of the non-traditional hydrodynamic shocks are not fully understood. Consequently, reliable and highly accurate numerical schemes for multiscale MHD equations pose a great challenge to algorithm development. In addition, controlling the numerical error of the divergence free condition of the magnetic fields for high order methods has been a stumbling block. Lower order methods are not practical for the astrophysical problems in question. We propose to extend our hydrodynamics schemes to the MHD equations with several desired properties over commonly used MHD schemes.