SEARCH · Search NASA
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.
Use of calculation program START and modeling program P 3D to produce radiation patterns of antennas mounted on a space station is discussed. Basic components of two space stations in the early design stage are simulated and radiation patterns for antennas mounted on the modules are presented.
A new type of corner-reflector mixer mount, which has the advantages of ease of fabrication and assembly as well as frequency versatility, has been designed and constructed. The predicted response patterns have been experimentally verified for various antenna lengths and operating frequencies between 800 and 2000 GHz. An important design feature is the incorporation of a microstrip matching network which eliminates IF impedance mismatch and provides mechanical isolation for the whisker antenna.
Large Language Models (LLMs) have demonstrated remarkable performance in various natural language processing tasks. However, the training of these models is computationally intensive and susceptible to faults, particularly in the attention mechanism, which is a critical component of transformer-based LLMs. In this paper, we investigate the impact of faults on LLM training, focusing on INF, NaN, and near-INF values in the computation results with systematic fault injection experiments. We observe the propagation patterns of these errors, which can trigger non-trainable states in the model and disrupt training, forcing the procedure to load from checkpoints. To mitigate the impact of these faults, we propose ATTNChecker, the first Algorithm-Based Fault Tolerance (ABFT) technique tailored for the attention mechanism in LLMs. ATTNChecker is designed based on fault propagation patterns of LLM and incorporates performance optimization to adapt to both system reliability and model vulnerability while providing lightweight protection for fast LLM training. Evaluations on four LLMs show that ATTNChecker on average incurs on average 7% overhead on training while detecting and correcting all extreme errors. Compared with the state-of-the-art checkpoint/restore approach, ATTNChecker reduces recovery overhead by up to 49×.
The basic mechanism underlying the generation of Moire patterns in line scanner data acquisition systems is examined. A general expression is developed in terms of typical system parameters for the reproduced image of such systems and the interaction of the image spectrum; the raster frequency and digital sampling frequency of the A/D conversion process are discussed and examples given. Systems design requirements for avoiding Moire pattern generation and two-dimensional aliasing are discussed.
The basic mechanism underlying the generation of Moire patterns in line scanner data acquisition systems is examined. A general expression is developed in terms of typical system parameters for the reproduced image of such systems and the interaction of the image spectrum; the raster frequency and digital sampling frequency of the A/D conversion process are discussed and examples given. System design requirements for avoiding Moire pattern generation and two-dimensional aliasing are discussed.
When gas and liquid mixtures flow in a pipe, the distribution of the two phases may take many forms. A flow pattern, or flow regime, is the characteristic spatial distribution of the phases of flow in a pipe. Because heat transfer and pressure drop are dependent on the characteristic distribution of phases, it is necessary to describe flow patterns in an appropriate manner so that a hydrodynamic or heat transfer theory applicable to that can be chosen. A theoretical two phase flow regime transition map under a microgravity environment was developed on physical concepts. These transitions use four basic flow patterns: dispersed flow, slug flow, stratified flow, and annular flow. The forces considered are body force, surface tension force, inertial force, friction, and the force from eddy turbulent fluctuation. Three dimensionless parameters were developed. Because these transition boundaries were developed based on physical concepts, they should be applicable to flow regimes occurring in various design conditions. Because the flow pattern data from KC-135 experiments are insufficient to verify these theoretical transition lines completely, an adiabatic experiment for flow regime analysis is recommended.
A waveguide feed has been designed to provide specified illumination patterns for a dual-reflector antenna in two wavelength bands: 8 to 9 GHz and 30 to 40 GHz. The feed (see figure) has a coaxial configuration: A wider circular tube surrounds a narrower circular tube that serves as a waveguide for the signals in the 30-to-40-GHz band. The annular space between the narrower and the wider tube serves as a coaxial waveguide for the signals in the 8-to-9-GHz band. The nominal design frequencies of the outer and inner waveguides are 8.45 and 32 GHz, respectively. Each of the two waveguides is terminated in a component that is sized and shaped to help focus the radiation in its respective frequency band into the specified illumination pattern. For the outer waveguide, the beam-shaping termination is a corrugated horn; for the inner waveguide, the beam-shaping termination is a dielectric rod insert.
Recommendations are made for upgrading two teleoperator/robotics test and simulation systems based upon a review of latest technology advances in the involved disciplines. A second generation Free Flying Mobility Unit is recommended which adds a sixth degree of freedom and incorporates other improvements which greatly expand the center's capability to perform evaluation tests and demonstrations of advanced systems concepts for rendezvous and docking in support of the Teleoperator Maneuvering System (TMS) Program. The Orbital Servicer System provides the capability for testing and demonstrating concepts for on orbit servicing of compatibly designed satellites/payloads. The TMS is to be the transporting vehicle for the servicer. The manipulator arm of the Orbital service System is presently computer controlled in the trajectory portion of the module transfer operation. The ultimate objective is to fully automte its operation requiring additional capabilities in sensors, artificial intelligence, image analysis, communications, computer programming, pattern recognition, kinematics, and manipulator design. It is recommended that the Electronics and Control Laboratory move to acquire the basic competencies in robotics necessary to achieve full automation.
This paper describes various possibilities for designing multibeam antennas using a single metasurface (MTS) aperture. Both single-source and multi-source feeding schemes are considered. For the single-source case, two approaches are investigated: i) division of the aperture in several angular sectors (one per beam) and ii) superposition of the individual modulations required to obtain the beams in the desired directions. A configuration based on a multi-source feeding scheme is also tailored by a superposition of modulation patterns. Numerical results based on the Method of Moments are presented for validation.
Mission Design and Navigation (MDNav) is an intensive process requiring advanced computational resources, expert human intuition, and many successive human-in-the-loop iterations to converge on acceptable trajectory designs or navigation solutions. The current bottleneck in MDNav is not the underlying computational algorithms but the human cognitive capacity to prune through a multitude of simulated results to select high-value candidates. One approach to alleviate this burden is through the judicious application of visualizations that allow humans to interactively filter data in multiple dimensions to reveal salient patterns and highlight divergences. When designed efficiently, such interactive visualizations should aid human operators to get familiar with data faster, visually observe correlations, and communicate findings more effortlessly. In this work, we present three visualization case studies that have the potential to increase human operator efficiency in MDNav. While identifying the most critical “pain points” that operators face, and also working on potential solutions, we followed a human-centered design approach. We started with a series of interviews with potential users, and then rapidly created prototypes for alternative solutions, validated outcomes with feedback from users through out development of these proof of concept visualizations. With this survey of our current efforts, we demonstrate the transformative capability of interactive data visualizations for improving mission development and operations, enabling operators to grow intuition, and communicating key concepts across diverse mission teams.
Flight experiment was designed primarily to determine the extent to which steady-state maneuvers influence characteristic vibration patterns measured at the input pinion and output annulus gear locations of the main transmission. If results were to indicate that maneuvers systematically influence vibration patterns, more extensive studies would be planned to explore the response surface. It was also designed to collect baseline data for comparison with experimental data to be recorded at a later date from test stands at Glenn Research Center. Finally, because this was the first vibration flight study on the Cobra aircraft, considerable energy was invested in developing an in-flight recording apparatus, as well as exploring acceleration mounting methods, and generally learning about the overall vibratory characteristics of the aircraft itself.
Various approaches to three-dimensional vision in space are reviewed with emphasis on the redundant 3D vision system designed for the Center for Intelligent Robotic Systems for Space Exploration. The system uses a controllable subset of five cameras, programmable structured light patterns, and sophisticated calibration routines. The design emphasizes real-time operation, human supervisory intervention, and the use of 3D vision to enhance the performance of cooperating robotic arms. Two methods of estimating the location of a point using 3D vision are discussed.
A design modification and a fabrication process that implements the modification have been conceived to solve two problems encountered in the development of back-illuminated, back-sidethinned complementary metal oxide/ semiconductor (CMOS) image-detector integrated circuits. The two problems are (1) how to form metal electrical-contact pads on the back side that are electrically connected through the thickness in proper alignment with electrical contact points on the front side and (2) how to provide alignment keys on the back side to ensure proper registration of backside optical components (e.g., microlenses and/or color filters) with the front-side pixel pattern. The essence of the design modification is to add metal plugs that extend from the desired front-side locations through the thickness and protrude from the back side of the substrate. The plugs afford the required front-to-back electrical conduction, and the protrusions of the plugs serve as both the alignment keys and the bases upon which the back-side electrical-contact pads can be formed.
While the causes of failures in complex engineered systems are often clear in hindsight, it can be challenging to predict failures proactively during the design of novel engineered products or systems. Identifying patterns can be useful for capturing common characteristics that may lead to failure. In this paper, we present a methodology for identifying patterns of failure from NASA’s publicly available Lessons Learned Information System (LLIS). We apply an ontology development and clustering approach to identify representative patterns leading to failures in historical lessons learned. A joint inductive-deductive approach reveals the key themes in lessons that lead to failure, which are formalized and recorded as an ontology of complex systems failure causes. Documents from the LLIS are manually tagged with relevant characteristics from the ontology. From the tagged set, clustering is used to capture co-occurring sets of characteristics that lead to failure. The primary contribution of this work is a method for extracting a set of generic failure patterns in complex engineered systems and characteristics for these patterns that can be identified at design time, knowledge of which can be used to plan mitigation strategies.
In large area, low cost solar cells of any type, the contact and grid structure metallization is an important factor which has an effect on the efficiency of the solar cell and its reliability. The present investigation is concerned with aspects of solar cell efficiency. An optimized metallization design leads to minimum total power loss, which is related to a simultaneous minimization of ohmic voltage drops and of shading of the front surface of the cell by the overlaid metal. The requirements regarding the design for a low-loss metallization pattern for the front surface of large area solar cells are represented by a set of design rules listed in a table. The total shading and voltage drop on such cells can be held to about 5%. However, not every metallization process is suited for meeting the requirements of the low-cost design. The low losses can be achieved only by use of several bus lines containing a bulk conductor, such as a wire.
While the causes of failures in complex engineered systems are often clear in hindsight, it can be challenging to predict failures proactively during the design of novel engineered products or systems. Identifying patterns can be useful for capturing common characteristics that may lead to failure. In this paper, we present a methodology for identifying patterns of failure from NASA’s publicly available Lessons Learned Information System (LLIS). We apply an ontology development and clustering approach to identify representative patterns leading to failures in historical lessons learned. A joint inductive-deductive approach reveals the key themes in lessons that lead to failure, which are formalized and recorded as an ontology of complex systems failure causes. Documents from the LLIS are manually tagged with relevant characteristics from the ontology. From the tagged set, clustering is used to capture co-occurring sets of characteristics that lead to failure. The primary contribution of this work is a method for extracting a set of generic failure patterns in complex engineered systems and characteristics for these patterns that can be identified at design time, knowledge of which can be used to plan mitigation strategies.
Electrochemical CO 2 reduction using renewable sources of electrical energy holds promise for converting CO 2 into fuels and chemicals. The complex interactions among chemical/electrochemical reactions and mass transport make it difficult to analyze the effect of an individual process on electrode performance based only on experimental methods. Here, we developed a generalized steady-state simulation to describe an electrode surface in which sequential cascade catalysts are patterned in a periodic trench design. If appropriately constructed, this trench geometry is hypothesized to be able to yield a higher net current density for a CO 2 reduction (CO 2 R) cascade reaction. We have used realistic experimental reaction kinetics to investigate the role of trench geometry in mass transport, local microenvironments, and selectivity for a model CO 2 R cascade reaction. The model considers local concentration gradients of bicarbonate species at quasi-equilibrium and catalytic surface reactions based on concentration-dependent Butler–Volmer kinetics. Our results suggest that varying the spatial distribution of active sites plays a significant role in facilitating effective mass transport between active sites, modulating selectivity for the cascade reaction, and enhancing the yield of desirable cascade products. Moreover, we observe that this trench geometry significantly alters the cascade reaction rate by affecting the local pH, which can cause inadvertent depletion of available aqueous CO 2 to limit the CO 2 R cascade kinetics and modest suppression of the hydrogen evolution reaction (HER). The results highlight the trade-offs between mass transport, pH, and reaction kinetics that become apparent only when considering the coupled physics of all processes at the electrode surface. Here, this model can thus serve as a primary tool to build more selective and efficient patterned architectures for the CO 2 R cascade catalysis.
Package of computer programs helps designers with layout and graphics of photomasks. Photomasks are specifically useful to applications involving fine reproducibility, repetition, and fabrication on planar surfaces of materials, items fabricated from photomasks include circuit boards, magnetic bubble devices and integrated optic circuits.