Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “high performance computing”

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.

At least 775 records · Page 43

A three-dimensional finite-element thermal/mechanical analytical technique for high-performance traveling wave tubes

Current research in high-efficiency, high-performance traveling wave tubes (TWT's) has led to the development of novel thermal/mechanical computer models for use with helical slow-wave structures. A three-dimensional, finite element computer model and analytical technique used to study the structural integrity and thermal operation of a high-efficiency, diamond-rod, K-band TWT designed for use in advanced space communications systems. This analysis focused on the slow-wave circuit in the radiofrequency section of the TWT, where an inherent localized heating problem existed and where failures were observed during earlier cold compression, or 'coining' fabrication technique that shows great potential for future TWT development efforts. For this analysis, a three-dimensional, finite element model was used along with MARC, a commercially available finite element code, to simulate the fabrication of a diamond-rod TWT. This analysis was conducted by using component and material specifications consistent with actual TWT fabrication and was verified against empirical data. The analysis is nonlinear owing to material plasticity introduced by the forming process and also to geometric nonlinearities presented by the component assembly configuration. The computer model was developed by using the high efficiency, K-band TWT design but is general enough to permit similar analyses to be performed on a wide variety of TWT designs and styles. The results of the TWT operating condition and structural failure mode analysis, as well as a comparison of analytical results to test data are presented.

Shalkhauser, Kurt A.↗

Visualization of supercomputer simulations in physics

A description is given of the hardware and software tools and techniques in use at NASA's Numerical Aerodynamic Simulation Facility for visualization of computational fluid dynamics. The hardware consists of high-performance graphics workstations connected to the supercomputer with high-bandwidth lines, a frame buffer connected to the supercomputer with UltraNet, a digital video recording system, and film recorders. The software permits the scientist to view the three-dimensional scenes dynamically, to zoom into a region of interest, and to rotate his viewing position to study any region of interest in more detail. The software also provides automated animation and video recording of the scenes. The digital effects unit on the video system facilitates comparison of computer simulations with flight or wind tunnel experiments.

Watson, Val↗

Refined AFC-Enabled High-Lift System Integration Study

A prior trade study established the effectiveness of using Active Flow Control (AFC) for reducing the mechanical complexities associated with a modern high-lift system without sacrificing aerodynamic performance at low-speed flight conditions representative of takeoff and landing. The current technical report expands on this prior work in two ways: (1) a refined conventional high-lift system based on the NASA Common Research Model (CRM) is presented that is more representative of modern commercial transport aircraft in terms of stall characteristics and maximum Lift/Drag (L/D) ratios at takeoff and landing-approach flight conditions; and (2) the design trade space for AFC-enabled high-lift systems is expanded to explore a wider range of options for improving their efficiency. The refined conventional high-lift CRM (HL-CRM) concept features leading edge slats and slotted trailing edge flaps with Fowler motion. For the current AFC-enhanced high lift system trade study, the refined conventional high-lift system is simplified by substituting simply-hinged trailing edge flaps for the slotted single-element flaps with Fowler motion. The high-lift performance of these two high-lift CRM variants is established using Computational Fluid Dynamics (CFD) solutions to the Reynolds-Averaged Navier-Stokes (RANS) equations. These CFD assessments identify the high-lift performance that needs to be recovered through AFC to have the CRM variant with the lighter and mechanically simpler high-lift system match the performance of the conventional high-lift system. In parallel to the conventional high-lift concept development, parametric studies using CFD guided the development of an effective and efficient AFC-enabled simplified high-lift system. This included parametric trailing edge flap geometry studies addressing the effects of flap chord length and flap deflection. As for the AFC implementation, scaling effects (i.e., wind-tunnel versus full-scale flight conditions) are addressed, as are AFC architecture aspects such as AFC unit placement, number AFC units, operating pressures, mass flow rates, and steady versus unsteady AFC applications. These efforts led to the development of a novel traversing AFC actuation concept which is efficient in that it reduces the AFC mass flow requirements by as much as an order of magnitude compared to previous AFC technologies, and it is predicted to be effective in driving the aerodynamic performance of a mechanical simplified high-lift system close to that of the reference conventional high-lift system. Conceptual system integration studies were conducted for the AFC-enhanced high-lift concept applied to a NASA Environmentally Responsible Aircraft (ERA) reference configuration, the so-called ERA-0003 concept. The results from these design integration assessments identify overall system performance improvement opportunities over conventional high-lift systems that suggest the viability of further technology maturation efforts for AFC-enabled high lift flap systems. To that end, technical challenges are identified associated with the application of AFC-enabled high-lift systems to modern transonic commercial transports for future technology maturation efforts.

Hartwich, Peter M.↗

Shaded-Color Picture Generation of Computer-Defined Arbitrary Shapes

SHADE computer program generates realistic color-shaded pictures from computer-defined arbitrary shapes. Objects defined for computer representation displayed as smooth, color-shaded surfaces, including varying degrees of transparency. Results also used for presentation of computational results. By performing color mapping, SHADE colors model surface to display analysis results as pressures, stresses, and temperatures. NASA has used SHADE extensively in sign and analysis of high-performance aircraft. Industry should find applications for SHADE in computer-aided design and computer-aided manufacturing. SHADE written in VAX FORTRAN and MACRO Assembler for either interactive or batch execution.

Cozzolongo, J. V.↗

Computer architecture for efficient algorithmic executions in real-time systems: New technology for avionics systems and advanced space vehicles

Improvements and advances in the development of computer architecture now provide innovative technology for the recasting of traditional sequential solutions into high-performance, low-cost, parallel system to increase system performance. Research conducted in development of specialized computer architecture for the algorithmic execution of an avionics system, guidance and control problem in real time is described. A comprehensive treatment of both the hardware and software structures of a customized computer which performs real-time computation of guidance commands with updated estimates of target motion and time-to-go is presented. An optimal, real-time allocation algorithm was developed which maps the algorithmic tasks onto the processing elements. This allocation is based on the critical path analysis. The final stage is the design and development of the hardware structures suitable for the efficient execution of the allocated task graph. The processing element is designed for rapid execution of the allocated tasks. Fault tolerance is a key feature of the overall architecture. Parallel numerical integration techniques, tasks definitions, and allocation algorithms are discussed. The parallel implementation is analytically verified and the experimental results are presented. The design of the data-driven computer architecture, customized for the execution of the particular algorithm, is discussed.

Carroll, Chester C.↗

An MPA-IO interface to HPSS

This paper describes an implementation of the proposed MPI-IO (Message Passing Interface - Input/Output) standard for parallel I/O. Our system uses third-party transfer to move data over an external network between the processors where it is used and the I/O devices where it resides. Data travels directly from source to destination, without the need for shuffling it among processors or funneling it through a central node. Our distributed server model lets multiple compute nodes share the burden of coordinating data transfers. The system is built on the High Performance Storage System (HPSS), and a prototype version runs on a Meiko CS-2 parallel computer.

Jones, Terry↗

An MPI-IO Interface to HPSS

This paper describes an implementation of the proposed MPI-IO standard for parallel I/O. Our system uses third-party transfer to move data over an external network between the processors where it is used and the I/O devices where it resides. Data travels directly from source to destination, without the need for shuffling it among processors or funneling it through a central node. Our distributed server model lets multiple compute nodes share the burden of coordinating data transfers. The system is built on the High Performance Storage System (HPSS), and a prototype version runs on a Meiko CS-2 parallel computer.

Parallel Processing↗

High speed network issues in a distributed visualization application

The author examines the issues in creating an UltraNet TCP/IP socket-based distributed application. The visualization computation is performed on a Connection Machine, and the results are rendered using a Silicon Graphics 4D/320 VGX workstation. The UltraNet network facilitates high-bandwidth communication between the computers. Ideally, taking advantage of the UltraNet is no more complex than developing TCP/IP and UNIX BSD socket-type applications on a single machine. In practice, there are several problems in developing an application using the UltraNet. The author identifies potential problems and discusses techniques for overcoming them. The performance of UltraNet communication is measured and found to be 10 MB/s for SGI 4D/320 workstations.

Krystynak, John↗

Numerical flow field visualization

The application of computer-generated imagery to analyze numerical solutions of fluid flow is presented. High-performance graphics workstations, their current capabilities, and their use in flow field analysis are discussed. Interactive flow field computations about complex three-dimensional bodies applying computer-generated imagery are projected for the future.

Smith, Robert E.↗

A three-dimensional finite-element thermal/mechanical analytical technique for high-performance traveling wave tubes

Current research in high-efficiency, high-performance traveling wave tubes (TWT's) has led to the development of novel thermal/ mechanical computer models for use with helical slow-wave structures. A three-dimensional, finite element computer model and analytical technique used to study the structural integrity and thermal operation of a high-efficiency, diamond-rod, K-band TWT designed for use in advanced space communications systems. This analysis focused on the slow-wave circuit in the radiofrequency section of the TWT, where an inherent localized heating problem existed and where failures were observed during earlier cold compression, or 'coining' fabrication technique that shows great potential for future TWT development efforts. For this analysis, a three-dimensional, finite element model was used along with MARC, a commercially available finite element code, to simulate the fabrication of a diamond-rod TWT. This analysis was conducted by using component and material specifications consistent with actual TWT fabrication and was verified against empirical data. The analysis is nonlinear owing to material plasticity introduced by the forming process and also to geometric nonlinearities presented by the component assembly configuration. The computer model was developed by using the high efficiency, K-band TWT design but is general enough to permit similar analyses to be performed on a wide variety of TWT designs and styles. The results of the TWT operating condition and structural failure mode analysis, as well as a comparison of analytical results to test data are presented.

Bartos, Karen F.↗

Preliminary Evaluation of Nonlinear Effects on TCA Flutter

The objective of this study is to investigate the effect of nonlinear aerodynamics, especially at high angles-of-attack with leading-edge separation, on the TCA flutter properties at transonic speeds. In order to achieve that objective, flutter simulations with Navier-Stokes CFD must be performed. To this end, time-marching Navier-Stokes solutions are computed for the TCA wing/body configuration at high angles-of-attack in transonic flight regimes. The approach is to perform non-linear flutter calculations on the TCA at two angles-of-attack, the first one being a case with attached flow (a=2.8 degrees) and the second one being a high angle-of-attack case with a wing leading edge vortex (a=12.11 degrees). Comparisons of the resulting histories and frequency damping information for both angles-of-attack will evaluate the impact of high-alpha aerodynamics on flutter.

Arslan, Alan E.↗

AFC-Enabled Simplified High-Lift System Integration Study

The primary objective of this trade study report is to explore the potential of using Active Flow Control (AFC) for achieving lighter and mechanically simpler high-lift systems for transonic commercial transport aircraft. This assessment was conducted in four steps. First, based on the Common Research Model (CRM) outer mold line (OML) definition, two high-lift concepts were developed. One concept, representative of current production-type commercial transonic transports, features leading edge slats and slotted trailing edge flaps with Fowler motion. The other CRM-based design relies on drooped leading edges and simply hinged trailing edge flaps for high-lift generation. The relative high-lift performance of these two high-lift CRM variants is established using Computational Fluid Dynamics (CFD) solutions to the Reynolds-Averaged Navier-Stokes (RANS) equations for steady flow. These CFD assessments identify the high-lift performance that needs to be recovered through AFC to have the CRM variant with the lighter and mechanically simpler high-lift system match the performance of the conventional high-lift system. Conceptual design integration studies for the AFC-enhanced high-lift systems were conducted with a NASA Environmentally Responsible Aircraft (ERA) reference configuration, the so-called ERA-0003 concept. These design trades identify AFC performance targets that need to be met to produce economically feasible ERA-0003-like concepts with lighter and mechanically simpler high-lift designs that match the performance of conventional high-lift systems. Finally, technical challenges are identified associated with the application of AFC-enabled highlift systems to modern transonic commercial transports for future technology maturation efforts.

Hartwich, Peter M.↗

On the Floating Point Performance of the i860 Microprocessor

The i860 microprocessor is a pipelined processor that can deliver two double precision floating point results every clock. It is being used in the Touchstone project to develop a teraflop computer by the year 2000. With such high computational capabilities it was expected that memory bandwidth would limit performance on many kernels. Measured performance of three kernels showed performance is less than what memory bandwidth limitations would predict. This paper develops a model that explains the discrepancy in terms of memory latencies and points to some problems involved in moving data from memory to the arithmetic pipelines.

Lee, King↗

Installed F/A-18 inlet flow calculations at 30 degrees angle-of-attack: A comparative study

NASA Lewis is currently engaged in a research effort as a team member of the High Alpha Technology Program (HATP) within NASA. This program utilizes a specially equipped F/A-18, the High Alpha Research Vehicle (HARV), in an ambitious effort to improve the maneuverability of high-performance military aircraft at low subsonic speed, high angle of attack conditions. The overall objective of the Lewis effort is to develop inlet technology that will ensure efficient airflow delivery to the engine during these maneuvers. One part of the Lewis approach utilizes computational fluid dynamics codes to predict the installed performance of inlets for these highly maneuverable aircraft. Full Navier-Stokes (FNS) calculations on the installed F/A-18 inlet at 30 degrees angle of attack, 0 degrees yaw, and a freestream Mach number of 0.2 have been obtained in this study using an algebraic turbulence model with two grids (original and revised). Results obtained with the original grid were used to determine where further grid refinements and additional geometry were needed. In order to account properly for the external effects, the forebody, leading edge extension (LEX), ramp, and wing were included with inlet geometry. In the original grid, the diverter, LEX slot, and leading edge flap were not included due to insufficient geometry definition, but were included in a revised grid. In addition, a thin-layer Navier-Stokes (TLNS) code is used with the revised grid and the numerical results are compared to those obtained with the FNS code. The TLNS code was used to evaluate the effects on the solution using a code with more recent CFD developments such as upwinding with TVD schemes versus central differencing with artificial dissipation. The calculations are compared to a limited amount of available experimental data. The predicted forebody/fuselage surface static pressures compared well with data of all solutions. The predicted trajectory of the vortex generated under the LEX was different for each solution. These discrepancies are attributed to differences in the grid resolution and turbulence modeling. All solutions predict that this vortex is ingested by the inlet. The predicted inlet total pressure recoveries are lower than data and the distortions are higher than data. The results obtained with the revised grid were significantly improved from the original grid results. The original grid results indicated the ingested vortex migrated to the engine face and caused additional distortions to those already present due to secondary flow development. The revised grid results indicate that the ingested vortex is dissipated along the inlet duct inboard wall. The TLNS results indicate the flow at the engine face was much more distorted than the FNS results and is attributed to the pole boundary condition introducing numerical distortions into the flow field.

Smith, C. Frederic↗

Prediction of Wind Tunnel Effects on the Installed F/A-18A Inlet Flow Field at High Angles-of-attack

NASA Lewis is currently engaged in a research effort as a team member of the High Alpha Technology Program (HATP) within NASA. This program utilizes a specially equipped F/A-18A, the High Alpha Research Vehicle (HARV), in an ambitious effort to improve the maneuverability of high performance military aircraft at low subsonic speed, high angle of attack conditions. The overall objective of the Lewis effort is to develop inlet technology that will ensure efficient airflow delivery to the engine during these maneuvers. One part of the Lewis approach utilizes computational fluid dynamics codes to predict the installed performance of inlets for these highly maneuverable aircraft. Wind tunnel tests were a major component of the Lewis program. Since the available wind tunnel was small (9 x 15 ft) as compared to the scale of the model of the F/A-18A (19.78 percent), there were questions about the capability to obtain useful inlet performance data. The blockage effects were expected to be very large. This report represents the results of an analysis to determine how the wind tunnel walls effect inlet performance at several angles of attack. The predictions for the external particle traces along the fuselage indicate the influence of the wind tunnel side wall under the model is greater at 30 deg angle of attack than at 50 deg angle of attack on the under Leading Edge Extension (LEX) vortex trajectory. The side wall above the model appears to have negligible influence on the under LEX vortex. This may be due to the LEX acting as 'shield' to the upper wall effects. As expected, the wind tunnel has a significant influence on the external forces. The lift and drag coefficients increase significantly for the wind tunnel model as compared to free stream conditions. The wind tunnel had a small effect on the inlet recovery and on inlet total pressure distortion patterns. The predicted recoveries for the wind tunnel model are within one percentage point of the model recoveries in free stream conditions.

Smith, Crawford F.↗

Parallel language constructs for tensor product computations on loosely coupled architectures

Distributed memory architectures offer high levels of performance and flexibility, but have proven awkard to program. Current languages for nonshared memory architectures provide a relatively low level programming environment, and are poorly suited to modular programming, and to the construction of libraries. A set of language primitives designed to allow the specification of parallel numerical algorithms at a higher level is described. Tensor product array computations are focused on along with a simple but important class of numerical algorithms. The problem of programming 1-D kernal routines is focused on first, such as parallel tridiagonal solvers, and then how such parallel kernels can be combined to form parallel tensor product algorithms is examined.

Mehrotra, Piyush↗

Design and control of a macro-micro robot for precise force applications

Creating a robot which can delicately interact with its environment has been the goal of much research. Primarily two difficulties have made this goal hard to attain. The execution of control strategies which enable precise force manipulations are difficult to implement in real time because such algorithms have been too computationally complex for available controllers. Also, a robot mechanism which can quickly and precisely execute a force command is difficult to design. Actuation joints must be sufficiently stiff, frictionless, and lightweight so that desired torques can be accurately applied. This paper describes a robotic system which is capable of delicate manipulations. A modular high-performance multiprocessor control system was designed to provide sufficient compute power for executing advanced control methods. An 8 degree of freedom macro-micro mechanism was constructed to enable accurate tip forces. Control algorithms based on the impedance control method were derived, coded, and load balanced for maximum execution speed on the multiprocessor system. Delicate force tasks such as polishing, finishing, cleaning, and deburring, are the target applications of the robot.

Wang, Yulun↗

Computational Analysis of a Prototype Martian Rotorcraft Experiment

This paper presents Reynolds-averaged Navier-Stokes calculations for a prototype Martian rotorcraft. The computations are intended for comparison with an ongoing Mars rotor hover test at NASA Ames Research Center. These computational simulations present a new and challenging problem, since rotors that operate on Mars will experience a unique low Reynolds number and high Mach number environment. Computed results for the 3-D rotor differ substantially from 2-D sectional computations in that the 3-D results exhibit a stall delay phenomenon caused by rotational forces along the blade span. Computational results have yet to be compared to experimental data, but computed performance predictions match the experimental design goals fairly well. In addition, the computed results provide a high level of detail in the rotor wake and blade surface aerodynamics. These details provide an important supplement to the expected experimental performance data.

Corfeld, Kelly J.↗