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.
This repository implements an iterative magnitude pruning algorithm for pruning neural networks in PyTorch. The pruning method involves gradually removing less significant weights from the model to achieve a specified sparsity, followed by fine-tuning the pruned model to recover performance.
The Chemical Recommender System (CRS) is an open-source, high-performance toolkit that enables real-time similarity searches across the complete PubChem database (over 50 million molecules) using commodity hardware. The CRS addresses critical limitations in existing chemical informatics platforms through a novel vector database infrastructure, extensible model integration capabilities, and complete algorithmic transparency. The system implements a vector database deployment with partitioned indexing that achieves a ~60x speedup over traditional approaches. A containerized model integration framework allows researchers to seamlessly incorporate custom predictive models into the full-scale search and scoring pipeline, while complete configurability of search parameters, filtering logic, and scoring functions provides capabilities not available in existing black-box solutions. Beyond structural similarity, the CRS integrates OPERA QSAR models for thermophysical and toxicity predictions, RDKit synthetic accessibility scoring, and user-defined models to compute weighted final replacement scores. The complete system is accessible through an interactive web application supporting real-time progress monitoring, post-processing score re-weighting, automated PDF reporting, and batch processing capabilities.
Trustworthy and interpretable uncertainty quantification is a long-standing challenge in artificial intelligence. Simulation-based inference (SBI) comprises a broad swath of approaches for estimating latent parameters with uncertainties. Although flexible neural density estimators in SBI can be remark- ably expressive capturing highly structured, high-dimensional posteriors their credible regions can be badly mis-calibrated and are often only accompanied by heuristic coverage checks. We present the first SBI framework that delivers finite-sample local valid coverage guarantees that hold in the neighborhood of each observation. Our framework can couple any off-the-shelf hierarchical SBI engine with a confor- mal Bayesian post-processing step that operates on the posterior predictive density. A kernel-weighted conformity score adapts the conformal quantile to the local geometry of the data, yielding prediction sets that are simultaneously (i) marginally calibrated, (ii) locally valid, and (iii) hierarchical, handling global and observation-specific parameters in a single pass. Through experiments on synthetic data and benchmarks from neuroscience and physics, we show that our approach attains 1 − α coverage, where prior SBI methods under- or over-cover. Our approach also maintains a competitive, credible set size with minimal computational overhead. Finally, our approach can be used to make predictions on real data and give valid credible regions modulo weight-initialization-based model mis-specification.
Mathematical model to design weight optimized silicon-germanium solar flat plate thermoelectric generator as spacecraft auxiliary power source in near sun missions
The concept is described of deducing the temperature profile of a planetary atmosphere from orbiter measurements of the planet's IR limb radiance profile. Expressions are derived for the weighting functions associated with the limb radiance profile for two infrared tranmission models. Analysis of the weighting functions for the Martian atmosphere indicates that a limb radiance profile in the 15 micron CO2 band can be used to determine the Martian atmospheric temperature profile from 20 to 60 km.
The identification of technologies necessary for development of millimeter spectrum communication satellites was examined from a system point of view. Development of methodology based on the technical requirements of potential services that might be assigned to millimeter wave bands for identifying the viable and appropriate technologies for future NASA millimeter research and development programs, and testing of this methodology with selected user applications and services were the goals of the program. The entire communications network, both ground and space subsystems was studied. Cost, weight, and performance models for the subsystems, conceptual design for point-to-point and broadcast communications satellites, and analytic relationships between subsystem parameters and an overall link performance are discussed along with baseline conceptual systems, sensitivity studies, model adjustment analyses, identification of critical technologies and their risks, and brief research and development program scenarios for the technologies judged to be moderate or extensive risks. Identification of technologies for millimeter satellite communication systems, and assessment of the relative risks of these technologies, was accomplished through subsystem modeling and link optimization for both point-to-point and broadcast applications.
Payload carrier designed for space vehicles is essentially cargo carrier that supports, positions, and protects various equipment and materials used in conducting experiments in weightless space environment. Proposed carrier entitled Materials Experiment Assembly II (MEA-II) is considered superior to previously developed models in size, weight, and cost to user. Structure is lightweight with insulated exterior and can be custom sized to meet user needs.
The objectives of the program were: (1) development of methodology based on the technical requirements of potential services that might be assigned to millimeter wave bands for identifying the viable and appropriate technologies for future NASA millimeter research and development programs, and (2) testing of this methodology with user applications and services. The scope of the program included the entire communications network, both ground and space subsystems. The reports include: (1) cost, weight, and performance models for the subsystems, (2) conceptual design for point-to-point and broadcast communications satellites, (3) analytic relationships between subsystem parameters and an overall link performance, (4) baseline conceptual systems, (5) sensitivity studies, (6) model adjustment analyses, (7) identification of critical technologies and their risks, (8) brief R&D program scenarios for the technologies judged to be moderate or extensive risks.
The significant technology items of the Concorde and the conceptual MCD baseline advanced supersonic transport are compared. The four major improvements are in the areas of range performance, structures (materials), aerodynamics, and in community noise. Presentation charts show aerodynamic efficiency; the reoptimized wing; low scale lift/drag ratio; control systems; structural modeling and analysis; weight and cost comparisons for superplasticity diffusion bonded titanium sandwich structures and for aluminum brazed titanium honeycomb structures; operating cost reduction; suppressor nozzles; noise reduction and range; the bicone inlet; a market summary; environmental issues; high priority items; the titanium wing and fuselage test components; and technology validation.
Studies were conducted to establish the effects of specimen geometry on the thermo-oxidative stability and the mechanical properties retention of unidirectional Celion 12000 graphite fiber reinforced PMR-15 polyimide composites. Weight loss, flexural strength and interlaminar shear strength are measured at isothermal aging times as long as 1639 hr at a temperature of 316 C for three different specimen geometries. It is found that the three different types of specimen surfaces exhibit different values of weight loss/unit area. The mechanical properties retention is also found to be dependent on geometry for these composites. The interlaminar shear strength decreases significantly over the complete range of aging times. The flexural strength retention starts showing geometric dependency after about 1000 hr of aging at 316C. Weight loss fluxes, associated with the three different types of exposed surfaces, are calculated and used to develop an empirical mathematical model for predicting the weight loss behavior of unidirectional composites of arbitrary geometries. Data are presented comparing experimentally determined weight loss with weight loss values predicted using the empirical model.
Studies were conducted to establish the effects of specimen geometry on the thermo-oxidative stability and the mechanical properties retention of unidirectional Celion 12000 graphite fiber reinforced PMR-15 polyimide composites. Weight loss, flexural strength and interlaminar shear strength are measured at isothermal aging times as long as 1639 hr at a temperature of 316 C for three different specimen geometries. It is found that the three different types of specimen surfaces exhibit different values of weight loss/unit area. The mechanical properties retention is also found to be dependent on geometry for these composites. The interlaminar shear strength decreases significantly over the complete range of aging times. The flexural strength retention starts showing geometric dependency after about 1000 hr of aging at 316 C. Weight loss fluxes, associated with the three different types of exposed surfaces, are calculated and used to develop an empirical mathematical model for predicting the weight loss behavior of unidirectional composites of arbitrary geometries. Data are presented comparing experimentally determined weight loss with weight loss values predicted using the empirical model.
A plasma model consisting of warm downward streaming electrons, cool Boltzmann electrons, cold upward streaming ions, and hot Boltzmann ions is considered. It is shown that there is little difference between having a finite temperature in the electron beam and leaving the beam cold. This implies a greater utility for a cold electron beam model, and adds weight to the idea that the smaller amplitude electrostatic shocks seen in conjunction with ion beams are nonlinear ion acoustic modes.
The fitting and proper regression coefficients were made of one hundred seventeen 10 x 10' blocks with observed gravity data and corresponding elevation in the Taiwan Island. To compare five different predicted models, and the proper one for the mean gravity anomalies were determined. The predicted gravity anomalies of the non-observed gravity blocks were decided when the coefficients obtained through the model with the weighted mean method. It was suggested that the mean gravity anomalies of 10 x 10' blocks should be made when comprehensive the observed and predicted data.
General pointing error and variance estimators for an open-loop conical scan (conscan) system are derived and analyzed. The conscan algorithm is modeled as a weighted least-squares estimator whose inputs are samples of receiver carrier power and its associated measurement uncertainty. When the assumptions of constant measurement noise and zero pointing error estimation are applied, the variance equation is then strictly a function of the carrier power to uncertainty ratio and the operator selectable radius and period input to the algorithm. The performance equation is applied to a 34-m mirror-based beam-waveguide conscan system interfaced with the Block V Receiver Subsystem tracking a Ka-band (32-GHz) downlink. It is shown that for a carrier-to-noise power ratio greater than or equal to 30 dB-Hz, the conscan period for Ka-band operation may be chosen well below the current DSN minimum of 32 sec. The analysis presented forms the basis of future conscan work in both research and development as well as for the upcoming DSN antenna controller upgrade for the new DSS-24 34-m beam-waveguide antenna.
Many acceleration-sensitive, microgravity science experiments will require active vibration isolation from manned orbiters on which they will be mounted. The isolation problem, especially in the case of a tethered payload, is a complex three-dimensional one that is best suited to modern-control design methods. In this paper, extended H(sub 2) synthesis is used to design an active isolator (i.e., controller) for a realistic single-input-multiple-output (SIMO) microgravity vibration isolation problem. Complex mu-analysis methods are used to analyze the isolation system with respect to sensor, actuator, and umbilical uncertainties. The paper fully discusses the design process employed and the insights gained. This design case study provides a practical approach for isolation problems of greater complexity. Issues addressed include a physically intuitive state-space description of the system, disturbance and noise filters, filters for frequency weighting, and uncertainty models. The controlled system satisfies all the performance specifications and is robust with respect to model uncertainties.
The early breed of slow commercial airliners did not require high-lift systems because their wing loadings were low and their speed ratios between cruise and low speed (takeoff and landing) were about 2:1. However, even in those days the benefit of high-lift devices was recognized. Simple trailing-edge flaps were in use, not so much to reduce landing speeds, but to provide better glide-slope control without sideslipping the airplane and to improve pilot vision over the nose by reducing attitude during low-speed flight. As commercial-airplane cruise speeds increased with the development of more powerful engines, wing loadings increased and a real need for high-lift devices emerged to keep takeoff and landing speeds within reasonable limits. The high-lift devices of that era were generally trailing-edge flaps. When jet engines matured sufficiently in military service and were introduced commercially, airplane speed capability had to be increased to best take advantage of jet engine characteristics. This speed increase was accomplished by introducing the wing sweep and by further increasing wing loading. Whereas increased wing loading called for higher lift coefficients at low speeds, wing sweep actually decreased wing lift at low speeds. Takeoff and landing speeds increased on early jet airplanes, and, as a consequence, runways worldwide had to be lengthened. There are economical limits to the length of runways; there are safety limits to takeoff and landing speeds; and there are speed limits for tires. So, in order to hold takeoff and landing speeds within reasonable limits, more powerful high-lift devices were required. Wing trailing-edge devices evolved from plain flaps to Fowler flaps with single, double, and even triple slots. Wing leading edges evolved from fixed leading edges to a simple Krueger flap, and from fixed, slotted leading edges to two- and three-position slats and variable-camber (VC) Krueger flaps. The complexity of high-lift systems probably peaked on the Boeing 747, which has a VC Krueger flap and triple-slotted, inboard and outboard trailing-edge flaps. Since then, the tendency in high-lift system development has been to achieve high levels of lift with simpler devices in order to reduce fleet acquisition and maintenance costs. The intent of this paper is to: (1) review available high-lift devices, their functions, and design criteria; (2) appraise high-lift systems presently in service on commercial air liners; (3) present personal study results on high-lift systems; (4) develop a weight and cost model for high-lift systems; and (5) discuss the development tendencies of future high-lift systems.
An open cycle gas turbine numerical modelling code suitable for thermodynamic performance analysis (i.e. thermal efficiency, specific fuel consumption, cycle state points, working fluid flowrates etc.) of automotive and aircraft powerplant applications has been generated at the NASA Lewis Research Center's Power Technology Division. The use this code can be made available to automotive gas turbine preliminary design efforts, either in its present version, or, assuming that resources can be obtained to incorporate empirical models for component weight and packaging volume, in later version that includes the weight-volume estimator feature. The paper contains a brief discussion of the capabilities of the presently operational version of the code, including a listing of input and output parameters and actual sample output listings.
The ion optics of NASA's Ion Neutral Mass Spectrometer (INMS) sensor was simulated with three dimensional models of the open source, the quadrupole deflector, the exit lens system and the quadrupole mass analyzer to design more compact models with lower weight. Comparison of calculated transmission with experimental results shows good agreement. Transmission analyses with varying geometrical parameters and voltages throw light on possible ways of reducing the size of the sensor. Trajectories of ions of mass 1-99 amu were simulated to analyze and optimize transmission. Analysis of open source transmission with varying angle of attack shows that the angular acceptance can be considerably increased by programming the voltages on the ion trap/ collimator. Analysis of transmission sensitivity to voltages and misalignments of the quadrupole deflector rods indicate that increased transmission is possible with a geometrically asymmetrical deflector and a deflector can be designed with much lower sensitivities of transmission. Bringing the disks closer together can decrease the size of the quadrupole deflector and also increase transmission. The exit lens system can be redesigned to be smaller by eliminating at least one electrode entirely without loss of transmission. Ceramic materials were investigated to find suitable candidates for use in the construction of lighter weight mass spectrometer. A high-sensitivity, high-resolution portable gas chromatograph mass spectrometer with a mass range of 2-700 amu has been built and will be commercialized in Phase 3.