Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Randomized methods”

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 937 records · Page 52
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Deep-learning-guided high-resolution subsurface reflectivity imaging with application to ground-penetrating radar data

Subsurface reflectivity imaging is one of the most important geophysical characterization methods for revealing subsurface structures. In many cases, accurate subsurface reflectivity imaging is challenging because of, for example, random or coherent noise in the data and sparse source-receiver observation geometry. Here, we develop a deep-learning-guided iterative imaging method to improve subsurface structure imaging. Specifically, we train a supervised neural network to infer a noise-free, high-resolution image from a noisy, low-resolution image and use this estimated image as guidance to regularize least-squares imaging. We develop a systematic method to generate high-quality synthetic training data (data-label pairs) to train the guidance neural network. The trained neural network can provide high-fidelity predictions even for field-data images that are not in the training data. We validate our new imaging method using one synthetic and two field ground-penetrating radar data examples, and find that our method can produce clean, high-resolution subsurface reflectivity images where existing single-pass and least-squares imaging methods fail due to noise and insufficient data coverage.

58 GEOSCIENCES↗
NASA NTRS2003

Comparison of Structural Optimization Techniques for a Nuclear Electric Space Vehicle

The purpose of this paper is to utilize the optimization method of genetic algorithms (GA) for truss design on a nuclear propulsion vehicle. Genetic Algorithms are a guided, random search that mirrors Darwin s theory of natural selection and survival of the fittest. To verify the GA s capabilities, other traditional optimization methods were used to compare the results obtained by the GA's, first on simple 2-D structures, and eventually on full-scale 3-D truss designs.

Benford, Andrew↗
DOE OSTI2021

Qutrit Randomized Benchmarking

Ternary quantum processors offer significant potential computational advantages over conventional qubit technologies, leveraging the encoding and processing of quantum information in qutrits (three-level systems). Therefore, to evaluate and compare the performance of such emerging quantum hardware it is essential to have robust benchmarking methods suitable for a higher-dimensional Hilbert space. We demonstrate extensions of industry standard randomized benchmarking (RB) protocols, developed and used extensively for qubits, suitable for ternary quantum logic. Using a superconducting five-qutrit processor, we find an average single-qutrit process infidelity of 3.8×10 -3 . Through interleaved RB, we characterize a few relevant gates, and employ simultaneous RB to fully characterize crosstalk errors. Finally, we apply cycle benchmarking to a two-qutrit CSUM gate and obtain a two-qutrit process fidelity of 0.85. Our results present and demonstrate RB-based tools to characterize the performance of a qutrit processor, and a general approach to diagnose control errors in future qudit hardware.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1996

Uncertainty and Intelligence in Computational Stochastic Mechanics

Classical structural reliability assessment techniques are based on precise and crisp (sharp) definitions of failure and non-failure (survival) of a structure in meeting a set of strength, function and serviceability criteria. These definitions are provided in the form of performance functions and limit state equations. Thus, the criteria provide a dichotomous definition of what real physical situations represent, in the form of abrupt change from structural survival to failure. However, based on observing the failure and survival of real structures according to the serviceability and strength criteria, the transition from a survival state to a failure state and from serviceability criteria to strength criteria are continuous and gradual rather than crisp and abrupt. That is, an entire spectrum of damage or failure levels (grades) is observed during the transition to total collapse. In the process, serviceability criteria are gradually violated with monotonically increasing level of violation, and progressively lead into the strength criteria violation. Classical structural reliability methods correctly and adequately include the ambiguity sources of uncertainty (physical randomness, statistical and modeling uncertainty) by varying amounts. However, they are unable to adequately incorporate the presence of a damage spectrum, and do not consider in their mathematical framework any sources of uncertainty of the vagueness type. Vagueness can be attributed to sources of fuzziness, unclearness, indistinctiveness, sharplessness and grayness; whereas ambiguity can be attributed to nonspecificity, one-to-many relations, variety, generality, diversity and divergence. Using the nomenclature of structural reliability, vagueness and ambiguity can be accounted for in the form of realistic delineation of structural damage based on subjective judgment of engineers. For situations that require decisions under uncertainty with cost/benefit objectives, the risk of failure should depend on the underlying level of damage and the uncertainties associated with its definition. A mathematical model for structural reliability assessment that includes both ambiguity and vagueness types of uncertainty was suggested to result in the likelihood of failure over a damage spectrum. The resulting structural reliability estimates properly represent the continuous transition from serviceability to strength limit states over the ultimate time exposure of the structure. In this section, a structural reliability assessment method based on a fuzzy definition of failure is suggested to meet these practical needs. A failure definition can be developed to indicate the relationship between failure level and structural response. In this fuzzy model, a subjective index is introduced to represent all levels of damage (or failure). This index can be interpreted as either a measure of failure level or a measure of a degree of belief in the occurrence of some performance condition (e.g., failure). The index allows expressing the transition state between complete survival and complete failure for some structural response based on subjective evaluation and judgment.

Ayyub, Bilal M.↗
DOE OSTI2021

Nonparametric, data-based kernel interpolation for particle-tracking simulations and kernel density estimation

Traditional interpolation techniques for particle tracking include binning and convolutional formulas that use pre-determined (i.e., closed-form, parameteric) kernels. In many instances, the particles are introduced as point sources in time and space, so the cloud of particles (either in space or time) is a discrete representation of the Green’s function of an underlying PDE. As such, each particle is a sample from the Green’s function; therefore, each particle should be distributed according to the Green’s function. In short, the kernel of a convolutional interpolation of the particle sample “cloud” should be a replica of the cloud itself. This idea gives rise to an iterative method by which the form of the kernel may be discerned in the process of interpolating the Green’s function. When the Green’s function is a density, this method is broadly applicable to interpolating a kernel density estimate based on random data drawn from a single distribution. We formulate and construct the algorithm and demonstrate its ability to perform kernel density estimation of skewed and/or heavy-tailed data including breakthrough curves.

42 ENGINEERING↗
DOE OSTI2024

Controlling radioisotope proportions when randomly sampling from Dirichlet distributions in PyRIID

As machine learning models for radioisotope quantification become more powerful, likewise the need for high-quality synthetic training data grows as well. For problem spaces that involve estimating the relative isotopic proportions of various sources in gamma spectra it is necessary to generate training data that accurately represents the variance of proportions encountered. In this report, we aim to provide guidance on how to target a desired variance of proportions which are randomly when using the PyRIID Seed Mixer, which samples from a Dirichlet distribution. We provide a method for properly parameterizing the Dirichlet distribution in order to maintain a constant variance across an arbitrary number of dimensions, where each dimension represents a distinct source template being mixed. We demonstrate that our method successfully parameterizes the Dirichlet distribution to target a specific variance of proportions, provided that several conditions are met. This allows us to follow a principled technique for controlling how random mixture proportions are generated which are then used downstream in the synthesis process to produce the final, noisy gamma spectra.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
NASA NTRS1993

On the apparent insignificance of the randomness of flexible joints on large space truss dynamics

Deployable periodic large space structures have been shown to exhibit high dynamic sensitivity to period-breaking imperfections and uncertainties. These can be brought on by manufacturing or assembly errors, structural imperfections, as well as nonlinear and/or nonconservative joint behavior. In addition, the necessity of precise pointing and position capability can require the consideration of these usually negligible and unknown parametric uncertainties and their effect on the overall dynamic response of large space structures. This work describes the use of a new design approach for the global dynamic solution of beam-like periodic space structures possessing parametric uncertainties. Specifically, the effect of random flexible joints on the free vibrations of simply-supported periodic large space trusses is considered. The formulation is a hybrid approach in terms of an extended Timoshenko beam continuum model, Monte Carlo simulation scheme, and first-order perturbation methods. The mean and mean-square response statistics for a variety of free random vibration problems are derived for various input random joint stiffness probability distributions. The results of this effort show that, although joint flexibility has a substantial effect on the modal dynamic response of periodic large space trusses, the effect of any reasonable uncertainty or randomness associated with these joint flexibilities is insignificant.

Koch, R. M.↗
NASA NTRS2020

Amino Acid Encoding for Deep Learning Applications

Background: The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-of-the-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results: By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion: Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Deep-learning↗
NASA NTRS2020

Amino Acid Encoding for Deep Learning Applications

Background: The number of applications of deep learning algorithms in bioinformatics is increasing as they usually achieve superior performance over classical approaches, especially, when bigger training datasets are available. In deep learning applications, discrete data, e.g. words or n-grams in language, or amino acids or nucleotides in bioinformatics, are generally represented as a continuous vector through an embedding matrix. Recently, learning this embedding matrix directly from the data as part of the continuous iteration of the model to optimize the target prediction – a process called ‘end-to-end learning’ – has led to state-ofthe-art results in many fields. Although usage of embeddings is well described in the bioinformatics literature, the potential of end-to-end learning for single amino acids, as compared to more classical manually-curated encoding strategies, has not been systematically addressed. To this end, we compared classical encoding matrices, namely one-hot, VHSE8 and BLOSUM62, to end-to-end learning of amino acid embeddings for two different prediction tasks using three widely used architectures, namely recurrent neural networks (RNN), convolutional neural networks (CNN), and the hybrid CNN-RNN. Results: By using different deep learning architectures, we show that end-to-end learning is on par with classical encodings for embeddings of the same dimension even when limited training data is available, and might allow for a reduction in the embedding dimension without performance loss, which is critical when deploying the models to devices with limited computational capacities. We found that the embedding dimension is a major factor in controlling the model performance. Surprisingly, we observed that deep learning models are capable of learning from random vectors of appropriate dimension. Conclusion: Our study shows that end-to-end learning is a flexible and powerful method for amino acid encoding. Further, due to the flexibility of deep learning systems, amino acid encoding schemes should be benchmarked against random vectors of the same dimension to disentangle the information content provided by the encoding scheme from the distinguishability effect provided by the scheme.

Hesham ElAbd↗
NASA NTRS2014

The Effect of General Statistical Fiber Misalignment on Predicted Damage Initiation in Composites

A micromechanical method is employed for the prediction of unidirectional composites in which the fiber orientation can possess various statistical misalignment distributions. The method relies on the probability-weighted averaging of the appropriate concentration tensor, which is established by the micromechanical procedure. This approach provides access to the local field quantities throughout the constituents, from which initiation of damage in the composite can be predicted. In contrast, a typical macromechanical procedure can determine the effective composite elastic properties in the presence of statistical fiber misalignment, but cannot provide the local fields. Fully random fiber distribution is presented as a special case using the proposed micromechanical method. Results are given that illustrate the effects of various amounts of fiber misalignment in terms of the standard deviations of in-plane and out-of-plane misalignment angles, where normal distributions have been employed. Damage initiation envelopes, local fields, effective moduli, and strengths are predicted for polymer and ceramic matrix composites with given normal distributions of misalignment angles, as well as fully random fiber orientation.

Micromechanical↗
DOE OSTI2022

Detecting hidden transient events in noisy nonlinear time-series

The information impulse function (IIF), running Variance, and local Hölder Exponent are three conceptually different time-series evaluation techniques. These techniques examine time-series for local changes in information content, statistical variation, and point-wise smoothness, respectively. Using simulated data emulating a randomly excited nonlinear dynamical system, this study interrogates the utility of each method to correctly differentiate a transient event from the background while simultaneously locating it in time. Computational experiments are designed and conducted to evaluate the efficacy of each technique by varying pulse size, time location, and noise level in time-series. Our findings reveal that, in most cases, the first instance of a transient event is more easily observed with the information-based approach of IIF than with the Variance and local Hölder Exponent methods. While our study highlights the unique strengths of each technique, the results suggest that very robust and reliable event detection for nonlinear systems producing noisy time-series data can be obtained by incorporating the IIF into the analysis.

42 ENGINEERING↗
NASA NTRS1981

Radiative transfer theory for passive microwave remote sensing of random media

The radiative transfer theory is applied to a random medium for the case of passive remote sensing. A general numerical solution with Gaussian quadrature method is studied. For a medium with laminar structure or a medium with cylindrical structure, closed form solutions are obtained by solving the radiative transfer equations of a two-layer random medium. Numerical results for the general case and the two limiting cases are compared.

Chuang, S. L.↗
NASA NTRS2015

Optimization of Microphone Locations for Acoustic Liner Impedance Eduction

Two impedance eduction methods are explored for use with data acquired in the NASA Langley Grazing Flow Impedance Tube. The first is an indirect method based on the convected Helmholtz equation, and the second is a direct method based on the Kumaresan and Tufts algorithm. Synthesized no-flow data, with random jitter to represent measurement error, are used to evaluate a number of possible microphone locations. Statistical approaches are used to evaluate the suitability of each set of microphone locations. Given the computational resources required, small sample statistics are employed for the indirect method. Since the direct method is much less computationally intensive, a Monte Carlo approach is employed to gather its statistics. A comparison of results achieved with full and reduced sets of microphone locations is used to determine which sets of microphone locations are acceptable. For the indirect method, each array that includes microphones in all three regions (upstream and downstream hard wall sections, and liner test section) provides acceptable results, even when as few as eight microphones are employed. The best arrays employ microphones well away from the leading and trailing edges of the liner. The direct method is constrained to use microphones opposite the liner. Although a number of arrays are acceptable, the optimum set employs 14 microphones positioned well away from the leading and trailing edges of the liner. The selected sets of microphone locations are also evaluated with data measured for ceramic tubular and perforate-over-honeycomb liners at three flow conditions (Mach 0.0, 0.3, and 0.5). They compare favorably with results attained using all 53 microphone locations. Although different optimum microphone locations are selected for the two impedance eduction methods, there is significant overlap. Thus, the union of these two microphone arrays is preferred, as it supports usage of both methods. This array contains 3 microphones in the upstream hard wall section, 14 microphones opposite the liner, and 3 microphones in the downstream hard wall section.

Jones, M. G.↗
NASA NTRSDate not supplied

Investigating the Effects of Exposure to Blue-Enriched Light or Peppermint Odor on Alertness, Mood, and Performance Upon Awakening from Deep Sleep at Night

Introduction: Sleep inertia refers the transient neurobehavioral impairments experienced immediately after waking from sleep. This period of reduced alertness and performance poses a significant safety risk to on-call workers who may be required to perform a safety-critical task immediately after waking (e.g., emergency services, health care, and military). In these operations, the need for a rapid return to full alertness is critical to mission safety and success. Several factors may exacerbate sleep inertia, resulting in greater impairment upon waking, including: waking from deep sleep, (i.e., slow wave sleep, SWS), waking at night, and waking following prior sleep loss. Awakenings under these conditions are common for on-call and extended shift workers who may need to perform safety-critical tasks soon after waking from unprotected sleep opportunities. Therefore, there is a need for evidence-based reactive countermeasures (i.e., used upon waking) to the cognitive consequences sleep inertia. Specifically, countermeasures that can rapidly restore alertness and performance immediately following sleep. A recent review of the literature on reactive countermeasures highlighted several research gaps and promising candidates for further investigation. The review also emphasized the need for countermeasures that are operationally viable and readily deployed in occupational settings. This study aims to address the identified gaps and limitations by assessing the efficacy of exposure to two known acute alerting stimuli - blue-enriched light and peppermint odor - to improve cognitive performance, alertness, and mood immediately after waking from SWS at night. Materials and Methods: Twelve participants completed a two-week within-subject, randomized, cross-over intervention study including two in-laboratory overnight visits. During each experimental week, the subjects experienced one intervention (light or peppermint) and a control condition upon awakening from SWS at night. The presentation order of the two conditions (intervention or control) at wake-up and the order of intervention (light or peppermint) by week was randomized by sex. Prior to each in-laboratory visit, participants maintained a sleep schedule of 8.5 h for 5 nights and 5 h for one night. Compliance with this sleep schedule was confirmed by actigraphy. In the laboratory, participants went to bed at their habitual bedtime and were monitored by standard polysomnography. After at least five minutes of continuous SWS, participants were awoken and exposed, in a randomized order, to either the control or intervention condition. During the hour after awakening from SWS (at 2, 17, 32, and 47 minutes after waking), participants completed a battery of tasks including a 5-minute psychomotor vigilance task (PVT), a subjective scale of alertness (Karolinska Sleepiness Scale, KSS), and visual analogue scales (VAS) of mood. Following this sleep inertia measurement period, all lights were turned off and participants were allowed to return to sleep. They were then awoken again from their subsequent SWS period and exposed to the alternative condition (control or intervention). Following this second awakening, participants were allowed to sleep until their habitual wake time and were then released from the laboratory. Participants then followed the at-home sleep schedule and returned to the laboratory for the second intervention (light or peppermint) following the procedures described above. The light intervention involved exposure to a blue-enriched light canvas illuminated for 1 hour at a distance of ~56 cm from the participant (~200 lux and ~60 melanopic lux at angle of gaze). For the peppermint intervention, peppermint oil was pipetted onto a mask, and participants inhaled the odor with the mask covering the nose and mouth for 1 minute. The control condition for both weeks involved a dim, red ambient light (<1 lux). An odorless mask, without any oil pipetted onto the mask, was also worn in the peppermint control condition. Results: Compared to the control condition, participants exposed to blue-enriched light had fewer PVT lapses (χ2 = 5.285, p = .022), reported feeling more alert (KSS: F1,77 = 4.955, p = .029; VASalert: F1,77 = 8.226, p = .005), and had improved mood (VAScheerful: F1,77 = 8.615, p = .004; VASdepressed: F1,77 = 4.649, p = .034; VASlethargic: F1,77 = 5.652, p = .020). Exposure to peppermint oil did not improve any outcome measures on any of the tasks compared to control condition (p > .05). Conclusions: We found that participants had fewer lapses of attention upon awakening when exposed to blue-enriched light compared to dim, red light. In addition, participants reported feeling more alert, more cheerful, less depressed, and less lethargic in the blue-enriched light condition. Brief exposure to a peppermint odor, however, did not appear to improve performance, alertness, or mood under the experimental conditions. Our null results in the peppermint condition may have been due to methodological limitations such as the duration and method of administration. Given the need to mitigate the potential impact of sleep inertia on safety-critical tasks in on-call operations, our findings suggest that blue-enriched light exposure upon awakening may help to improve performance and alertness during the sleep inertia period following awakening from deep, nocturnal sleep. We are currently exploring the potential mechanisms for the effect of light on cognitive performance upon awakening as well as investigating its application in real-world settings to explore the translational efficacy of this countermeasure to occupational environments. Continued exploration into light and other reactive countermeasures, and potentially their combination, is needed in order to provide evidence-based guidance on effective sleep inertia countermeasures to improve the alertness and performance of those required to perform safety-critical tasks soon after waking.

sleep inertia↗
DOE OSTI2021

Utilizing physics-based input features within a machine learning model to predict wind speed forecasting error

Machine learning is quickly becoming a commonly used technique for wind speed and power forecasting. Many machine learning methods utilize exogenous variables as input features, but there remains the question of which atmospheric variables are most beneficial for forecasting, especially in handling non-linearities that lead to forecasting error. This question is addressed via creation of a hybrid model that utilizes an autoregressive integrated moving-average (ARIMA) model to make an initial wind speed forecast followed by a random forest model that attempts to predict the ARIMA forecasting error using knowledge of exogenous atmospheric variables. Variables conveying information about atmospheric stability and turbulence as well as inertial forcing are found to be useful in dealing with non-linear error prediction. Streamwise wind speed, time of day, turbulence intensity, turbulent heat flux, vertical velocity, and wind direction are found to be particularly useful when used in unison for hourly and 3 h timescales. The prediction accuracy of the developed ARIMA–random forest hybrid model is compared to that of the persistence and bias-corrected ARIMA models. The ARIMA–random forest model is shown to improve upon the latter commonly employed modeling methods, reducing hourly forecasting error by up to 5 % below that of the bias-corrected ARIMA model and achieving an R 2 value of 0.84 with true wind speed.

17 WIND ENERGY↗
NASA NTRS1982

Investigation of continuously traversing microphone system for mode measurement

The continuously Traversing Microphone System consists of a data acquisition and processing method for obtaining the modal coefficients of the discrete, coherent acoustic field in a fan inlet duct. The system would be used in fan rigs or full scale engine installations where present measurement methods, because of the excessive number of microphones and long test times required, are not feasible. The purpose of the investigation reported here was to develop a method for defining modal structure by means of a continuously traversing microphone system and to perform an evaluation of the method, based upon analytical studies and computer simulated tests. A variety of system parameters were examined, and the effects of deviations from ideal were explored. Effects of traverse speed, digitizing rate, run time, roundoff error, calibration errors, and random noise background level were determined. For constant fan operating speed, the sensitivity of the method to normal errors and deviations was determined to be acceptable. Good recovery of mode coefficients was attainable. Fluctuating fan speed conditions received special attention, and it was concluded that by employing suitable time delay procedures, satisfactory information on mode coefficients can be obtained under realistic conditions. A plan for further development involving fan rig tests was prepared.

Cicon, D. E.↗
NASA NTRS1994

Fatigue reliability method with in-service inspections

The first reliability method (FORM) has traditionally been applied in a fatigue reliability setting to one inspection interval at a time, so that the random distribution of crack lengths must be recharacterized following each inspection. The FORM presented here allow each analysis to span several inspection periods without explicit characterization of the crack length distribution upon each inspection. The method thereby preserves the attractive feature of FORM in that relatively few realizations in the random variable space need to be considered. Examples are given which show that the present methodology gives estimates which are in good agreement with Monte Carlo simulations and is efficient even for complex components.

Harkness, H. H.↗
NASA NTRS1995

Microelectronics Instrument Products Shock and Vibration Electro-Optics: Acceptance Data Package - Section B

The motor/encoder (henceforth referred to as the UUT) test sequence began with a baseline functional evaluation, which demonstrated that the motor satisfied the operating torque, cogging torque, winding resistance, and mechanical requirements of SOW. In addition, the encoder electrical requirements were verified, as well as the alignment of the encoder outputs relative to the motor shaft position. There were no discrepancies observed in this portion of the test. The UUT was then exposed to a number of environments, including thermal vacuum, thermal cycling, random and sine vibration, and mechanical shock. During the thermal environments, the performance of the UUT under load was verified at specified points in the cycles, as described in ATP 20049. In addition, the UUT was bench tested between the two thermal environments. No anomalies were observed during the thermal tests. The vibration and shock tests were performed by East-West Technology Corporation, West Babylon, New York. The UUT was delivered to the lab in a sealed vibration fixture in order to maintain the cleanliness levels required by the SOW. In addition, a three ounce load was attached to the motor shaft. The attachment method of this load caused damage to the shaft and bearing during random vibration of S/N 0003 on April 28, 1995, and is described further in NCR 00168.

Source record↗