Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “unreliability”

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 757 records · Page 42

Sensing Small Uncrewed Aerial Vehicles with Distributed Radars for Advanced Air Mobility Surveillance

In the context of Advanced and Urban Air Mobility major attention is being reserved to the development of sensing strategies for small Uncrewed Aerial Vehicles (sUAVs) to enable their safe operations in and around urban areas. Such strategies should rely on non-cooperative distributed sensors to strengthen the surveillance solution towards the unreliability of Global Navigation Satellite System (GNSS) positioning information, which is typical for low-altitude-flying platforms in urban regions, and increase the monitored airspace volume. To this aim, this paper proposes a fusion solution for a network of distributed ground-based radars, which can be exploited to not only increase the coverage over large airspace volumes but also improve the overall detectability and traceability of sUAVs by leveraging on multiple views over the same area. The solution exploits a centralized fusion scheme in which measurements collected by each radar are shared with a Fusion Center where Kalman Filtering is exploited to build a unique, fused track. Tests conducted on experimental data collected using two sUAVs as flying targets and three distributed radars showed that the proposed solution can produce an increase in coverage from about 20 % (single radar configuration) to about 80 % of the targets’ flight path, as well as a finer accuracy yielding meter and meter-per-second root mean square error values on position and velocity components.

Federica Vitiello↗

The Lunar Explorer Instrument for Space Biology Applications (LEIA) Pre-Flight Tests: Mechanical Considerations​

The BioSensor payload on the upcoming LEIA platform aboard the CLPS (CP-22) lander will carry yeast to the Moon's south pole to study the effects of lunar radiation and gravity. From Earth, BioSensor will transport desiccated yeast within 16 fluidic cards, each containing 16 fluidic wells of 100 microliters volume. At the start of the experiment, these microbes will be autonomously rehydrated on the Moon, and optical absorbance measurements at three wavelengths will be performed across all 256 wells. Absorbance measurements in fluidic wells are susceptible to noise and artifacts due to bubble formation during well rehydration and microbial growth. BioSensor, as a biofluidics system, faces significant data anomalies caused by payload vibrations, particularly from the operation of the co-located PROSPECT’s ProSEED drill, to assess lunar resources by penetrating the subsurface. To evaluate the impact of drill vibrations, fully assembled and filled fluidic cards were subjected to sine sweep signals in the x, y, z planes, ranging from 5 - 2560 Hz at g-loads of 0.25 and 0.5, with on-board accelerometer for localized vibration measurement. Wells with various fill conditions, including partial fills with different bubble sizes and completely filled wells, were tested. Results indicated that small bubbles adhered to well walls at high frequencies but were dislodged at low frequencies, causing data anomalies. Medium and large bubbles exhibited higher interference with optical data due to lower surface tension and variable bubble surface flexion. Overall, it was found that irrespective of the fill type, low-frequency vibrations were most detrimental to optical readings. Understanding the drill schedule and correlating it with BioSensor timestamps will help identify periods when BioSensor data is unreliable, a crucial step for accurately interpreting the microbial response data and ensuring the validity of the BioSensor's scientific measurements on the lunar surface.

Chinmayee Govinda Raj↗

Swift Mission Gyro Patch and Re-Calibration Without A Calibration Campaign

The Swift spacecraft has three Two-Axis-Rate-Assemblies (TARAs) (designated as TARA 1, TARA 2, and TARA 3, and generically referred to as gyros) which until March of 2024 were all used in rate estimation as an operational workaround (dubbed “corrector-gyro”) of a software defect discovered in 2007. TARA 1 performance degraded throughout 2023 and 2024, making it an unreliable rate source. It was determined that shifting TARA 1 to be the corrector-gyro and using mainly TARAs 2 and 3, with the existing flight software (FSW) defect, would not meet performance requirements for science operations. Rather the FSW was patched to remove the defect found in 2007. This patch was delivered by the vendor in 2009, but was not fully tested or installed at that time. A decision was made in 2024 to install the patch despite limited simulation and testing capabilities. The problem facing the team was to come up with a set of alignment parameters that would work with the new patch, using data from the spacecraft operating normally in the old configuration. Calibration activities would have been possible in either the old or new FSW configuration, but were ultimately deemed unnecessary, saving staff time and pre-serving time on the spacecraft for science data collection. This paper describes the old and new FSW configurations related to gyro processing and alignment, and how the raw gyro and star tracker data from normal operation were used with a multidimensional unconstrained nonlinear minimization in MATLAB (Nelder-Mead1) to produce a fully calibrated set of gyro alignment parameters compatible with the patched FSW. Flight data is presented showing that not only did the process work, it improved performance significantly, rivaling the best slew performance of the mission to date.

MATLAB↗

Using support vector machines for anomalous change detonation

We cast anomalous change detection as a binary classification problem, and use a support vector machine (SVM) to build a detector that does not depend on assumptions about the underlying data distribution. To speed up the computation, our SVM is implemented, in part, on a graphical processing unit. Results on real and simulated anomalous changes are used to compare performance to algorithms which effectively assume a Gaussian distribution. In this paper, we investigate the use of support vector machines (SVMs) with radial basis kernels for finding anomalous changes. Compared to typical applications of SVMs, we are operating in a regime of very low false alarm rate. This means that even for relatively large training sets, the data are quite meager in the regime of operational interest. This drives us to use larger training sets, which in turn places more of a computational burden on the SVM. We initially considered three different approaches to to address the need to work in the very low false alarm rate regime. The first is a standard SVM which is trained at one threshold (where more reliable estimates of false alarm rates are possible) and then re-thresholded for the low false alarm rate regime. The second uses the same thresholding approach, but employs a so-called least squares SVM; here a quadratic (instead of a hinge-based) loss function is employed, and for this model, there are good theoretical arguments in favor of adjusting the threshold in a straightforward manner. The third approach employs a weighted support vector machine, where the weights for the two types of errors (false alarm and missed detection) are automatically adjusted to achieve the desired false alarm rate. We have found in previous experiments (not shown here) that the first two types can in some cases work well, while in other cases they do not. This renders both approaches unreliable for automated change detection. By contrast, the third approach reliably produces good results, but at the cost of larger computational requirements caused by the need to estimate very small false alarm rates. To address these computational requirements, we employ a recently developed in-house solver for SVMs that is significantly faster than freely available standard solvers. But these computational issues are secondary to the larger question: do kernelized solutions provide better performance, in terms of detection rates and false alarm rates, than more traditional methods for change detection that effectively assume Gaussian data distributions? To this end, we will compare ROC curves obtained from the SVM with those from chronochrome, covariance equalization, and hyperbolic anomalous change detection.

97 MATHEMATICS AND COMPUTING↗

Dynamic Time-Variant Connection Management for PGAS Models on InfiniBand

InfiniBand (IB) has established itself as a promising network infrastructure for high-end cluster computing systems as evidenced by its usage in the Top500 supercomputers today. While the IB standard describes multiple communication models (including reliable-connection (RC), and unreliable datagram (UD)), most of its promising features such as remote direct memory access (RDMA), hardware atomics and network fault tolerance are only available for the RC model which requires connections between communicating process pairs. In the past, several researchers have proposed on-demand connection management techniques that establish connections when there is a need to communicate, and not before. While such techniques work well for algorithms and applications that only communicate with a small set of processes in their life-time, there exists a broad set of applications that do not follow this trend. For example, applications that perform dynamic load balancing and adaptive work stealing have a small set of communicating neighbors at any given time, but over time the total number of neighbors can be very high; in some cases, equal to the entire system size. In this paper, we present a dynamic time-variant connection management approach that establishes connections on-demand like previous approaches, but further intelligently tears down some of the unused connections as well. While connection tear-down itself is relevant for any programming model, different models have different complexities. In this paper, we study the Global Arrays (GA) PGAS model for two reasons: (1) the simple one-sided communication primitives provided by GA and other PGAS models ensure that connection requests are always initiated by the origin process without explicit synchronization with the target process---this makes connection tear-down simpler to handle; and (2) GA supports several applications that demonstrate this behavior making it an obvious first target for the proposed enhancements. We evaluate our proposed approach using several micro-benchmarks as well as the NWChem computational chemistry application on more than 6000 processes, and show that our approach can significantly reduce the memory requirements of the communication library while maintaining its performance. To the best of our knowledge, this is the first design, implementation and evaluation of connection tear-down protocols over InfiniBand.

99 GENERAL AND MISCELLANEOUS↗

Modeling Impact-Induced Reactivity Changes Using DAG-MCNP5

There is a long literature studying the criticality of space reactors immersed in water/sand after a launch accident; however, most of these studies evaluate nominal or uniformly compacted system configurations. There is less research on the reactivity consequences of impact, which can cause large structural deformation of reactor components that can result in changes in the reactivity of the system. Predicting these changes is an important component of launch safety analysis. This paper describes new features added to the DAG-MCNP5 neutronics code that allow the criticality analysis of deformed geometries. A CAD-based solid model of the reactor geometry is used to generate an initial mesh for a structural mechanics impact calculation using the PRONTO3D/PRESTO continuum mechanics codes. Boundary conditions and material specifications for the reactivity analysis are attached to the solid model that is then associated with the initial mesh representation. This geometry is then updated with the deformed finite element mesh to perturb node coordinates. DAG-MCNP5 was extended to accommodate two consequences of the large structural deformations: dead elements representing fracture, and small overlaps between adjacent volumes. The dead elements are removed during geometry initialization and adjustments are made to conserve mass. More challenging, small overlaps where adjacent mesh elements contact cause the geometric queries to become unreliable. A new point membership test was developed that is tolerant of self-intersecting volumes, and the particle tracking algorithm was adjusted to enable transport through small overlaps. These new features enable DAG-MCNP5 to perform particle transport and criticality eigenvalue calculations on both deformed mesh geometry and CAD geometry with small geometric defects. Detailed impact simulations were performed on an 85-pin space reactor model. In the most realistic model that included NaK coolant and water in the impact simulation, the eigenvalue was determined to increase 2.7% due to impact.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Effect of numerical parameters on characterizing the hardening behavior of ductile uniaxial tension specimens.

Many problems of practical importance involve ductile materials that undergo very large strains, in many cases to the point of failure. Examples include structures subjected to impact or blast loads, energy absorbing devices subjected to significant crushing, cold-forming manufacturing processes and others. One of the most fundamental pieces of data that is required in the analysis of this kind of problems is the fit of the uniaxial stress-strain curve of the material. A series of experiments where mild steel plates were punctured with a conical indenter provided a motivation to characterize the true stress-strain curve until the point of failure of this material, which displayed significant ductility. The hardening curve was obtained using a finite element model of the tensile specimens that included a geometric imperfection in the form of a small reduction in the specimen width to initiate necking. An automated procedure iteratively adjusted the true stress-strain curve fit used as input until the predicted engineering stress-strain curve matched experimental measurements. Whereas the fitting is relatively trivial prior to reaching the ultimate engineering stress, the fit of the softening part of the engineering stress-stain curve is highly dependent on the finite element parameters such as element formulation and initial geometry. Results by two hexahedral elements are compared. The first is a standard, under-integrated, uniform-strain element with hourglass control. The second is a modified selectively-reduced-integration element. In addition, the effects of element size, aspect ratio and hourglass control characteristics are investigated. The effect of adaptively refining the mesh based on the aspect ratio of the deformed elements is also considered. The results of the study indicate that for the plate puncture problem, characterizing the material with the same element formulation and size as used in the plate models is beneficial. On the other hand, using different element formulations, sizes or initial aspect ratios can lead to unreliable results.

36 MATERIALS SCIENCE↗

Retrieving and Ranking Unannotated Images through Collaboratively Mining Online Search Results

We present a new image search and ranking algorithm for retrieving unannotated images by collaboratively mining online search results, which consist of online image and text search results. The online image search results are leveraged as reference examples to perform content-based image search over unannotated images. The online text search results are utilized to estimate individual reference images relevance to the search query as not all the online image search results are closely related to the query. Overall, the key contribution of our method lies in its capability to deal with unreliable online image search results through jointly mining visual and textual aspects of online search results. Through such collaborative mining, our algorithm infers the relevance of an online search result image to a text query. Once we estimate a query relevance score for each online image search result, we can selectively use query specific online search result images as reference examples for retrieving and ranking unannotated images. To explore the performance of our algorithm, we tested our algorithm both on the standard public image datasets and several modest sized personal photo collections. We also compared the performance of our method with that of two peer methods. The results are very positive, indicating that our algorithm is superior to existing content-based image search algorithms for retrieving and ranking unannotated images. Overall, the main advantage of our algorithm comes from its collaborative mining over online search results both in the visual and textual domains.

99 GENERAL AND MISCELLANEOUS↗

Droop Compensation for the High Voltage Converter Modulators at the Spallation Neutron Source.

The Spallation Neutron Source (SNS) has been in operation since 2006 and routinely operates at 1 MW of beam power. At 1 MW, the High Voltage Converter Modulators (HVCMs) are delivering 8-11 MW pulsed power to the Klystrons for up to 1185 us at 60 Hz. The pulsed output of the modulator has a substantial voltage droop operating in open-loop mode. Achieving 1.4 MW operations require an increase in the output pulse width of the HVCM, resulting in loss of LLRF control margin due to modulator voltage reduction at the end of the pulse. Initially, the HVCM was designed with a pulse width modulation scheme for droop compensation but early operations at full duty revealed this technique unreliable. Increasing the output voltage of the modulator would likely compromise system reliability. This paper proposes the use of alternate modulation schemes to address the voltage droop. The effect of frequency modulation and phase modulation on voltage droop is studied. Experimental results will also be presented.

43 PARTICLE ACCELERATORS↗

TAGGING, TRACKING AND LOCATING WITHOUT GPS

The Savannah River National Laboratory (SRNL) was requested to lead a Law Enforcement Working Group that was formed to collaborate on common operational needs. All agencies represented on the working group ranked their need to tag, track, and locate a witting or unwitting target as their highest priority. Specifically, they were looking for technologies more robust than Global Positioning Satellite (GPS), could communicate back to the owner, and worked where normal cell phone communications did not work or were unreliable. SRNL brought together multiple technologies in a demonstration that was held in in various Alaska venues, including metropolitan, wilderness, and at-sea that met the working group's requirements. Using prototypical technologies from Boeing, On Ramp, and Fortress, SRNL was able to demonstrate the ability to track personnel and material in all scenarios including indoors, in heavily wooden areas, canyons, and in parking garages. In all cases GPS signals were too weak to measure. Bi-directional communication was achieved in areas that Wi-Fi, cell towers, or traditional radios would not perform. The results of the exercise will be presented. These technologies are considered ideal for tracking high value material such has nuclear material with a platform that allows seamless tracking anywhere in the world, indoors or outdoors.

99 GENERAL AND MISCELLANEOUS↗

Efficient Solar Concentrators: Affordable Energy from Water and Sunlight

Broad Funding Opportunity Announcement Project: Teledyne is developing a liquid prism panel that tracks the position of the sun to help efficiently concentrate its light onto a solar cell to produce power. Typically, solar tracking devices have bulky and expensive mechanical moving parts that require a lot of power and are often unreliable. Teledyne’s liquid prism panel has no bulky and heavy supporting parts—instead it relies on electrowetting. Electrowetting is a process where an electric field is applied to the liquid to control the angle at which it meets the sunlight above and to control the angle of the sunlight to the focusing lensthe more direct the angle to the focusing lens, the more efficiently the light can be concentrated to solar panels and converted into electricity. This allows the prism to be tuned like a radio to track the sun across the sky and steer sunlight into the solar cell without any moving mechanical parts. This process uses very little power and requires no expensive supporting hardware or moving parts, enabling efficient and quiet rooftop operation for integration into buildings.

ARPA-E↗

Sodium-Beta Batteries for Grid-Scale Storage: Planar Sodium-Beta Batteries for Renewable Integration and Grid Applications

Broad Funding Opportunity Announcement Project: EaglePicher is developing a sodium-beta alumina (Na-Beta) battery for grid-scale energy storage. High-temperature Na-Beta batteries are a promising grid-scale energy storage technology, but existing approaches are expensive and unreliable. EaglePicher has modified the shape of the traditional, tubular-shaped Na-Beta battery. It is using an inexpensive stacked design to improve performance at lower temperatures, leading to a less expensive overall storage technology. The new design greatly simplifies the manufacturing process for beta alumina membranes (a key enabling technology), providing a subsequent pathway to the production of scalable, modular batteries at half the cost of the existing tubular designs.

ARPA-E↗

Evolutionary Tuning of Building Models to Monthly Electrical Consumption

Building energy models of existing buildings are unreliable unless calibrated so they correlate well with actual energy usage. Calibrating models is costly because it is currently an art which requires significant manual effort by an experienced and skilled professional. An automated methodology could significantly decrease this cost and facilitate greater adoption of energy simulation capabilities into the marketplace. The Autotune project is a novel methodology which leverages supercomputing, large databases of simulation data, and machine learning to allow automatic calibration of simulations to match measured experimental data on commodity hardware. This paper shares initial results from the automated methodology applied to the calibration of building energy models (BEM) for EnergyPlus (E+) to reproduce measured monthly electrical data.

Garrett, Aaron↗

Light-weight analyzer for odor recognition

The invention provides a light weight analyzer, e.g., detector, capable of locating clandestine graves. The detector utilizes the very specific and unique chemicals identified in the database of human decompositional odor. This detector, based on specific chemical compounds found relevant to human decomposition, is the next step forward in clandestine grave detection and will take the guess-work out of current methods using canines and ground-penetrating radar, which have historically been unreliable. The detector is self contained, portable and built for field use. Both visual and auditory cues are provided to the operator.

47 OTHER INSTRUMENTATION↗

Component Repair Times Obtained from MSPI Data

Information concerning times to repair or restore equipment to service given a failure is valuable to probabilistic risk assessments (PRAs). Examples of such uses in modern PRAs include estimation of the probability of failing to restore a failed component within a specified time period (typically tied to recovering a mitigating system before core damage occurs at nuclear power plants) and the determination of mission times for support system initiating event (SSIE) fault tree models. Information on equipment repair or restoration times applicable to PRA modeling is limited and dated for U.S. commercial nuclear power plants. However, the Mitigating Systems Performance Index (MSPI) program covering all U.S. commercial nuclear power plants provides up-to-date information on restoration times for a limited set of component types. This paper describes the MSPI program data available and analyzes the data to obtain median and mean component restoration times as well as non-restoration cumulative probability curves. The MSPI program provides guidance for monitoring both planned and unplanned outages of trains of selected mitigating systems deemed important to safety. For systems included within the MSPI program, plants monitor both train UA and component unreliability (UR) against baseline values. If the combined system UA and UR increases sufficiently above established baseline results (converted to an estimated change in core damage frequency or CDF), a “white” (or worse) indicator is generated for that system. That in turn results in increased oversight by the US Nuclear Regulatory Commission (NRC) and can impact a plant’s insurance rating. Therefore, there is pressure to return MSPI program components to service as soon as possible after a failure occurs. Three sets of unplanned outages might be used to determine the component repair durations desired in this article: all unplanned outages for the train type that includes the component of interest, only unplanned outages associated with failures of the component of interest, and only unplanned outages associated with PRA failures of the component of interest. The paper will describe how component repair times can be generated from each set and which approach is most applicable. Repair time information will be summarized for MSPI pumps and diesel generators using data over 2003 – 2007. Also, trend information over 2003 – 2012 will be presented to indicate whether the 2003 – 2007 repair time information is still considered applicable. For certain types of pumps, mean repair times are significantly higher than the typically assumed 24 h duration.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Method for training honeybees to respond to olfactory stimuli and enhancement of memory retention therein

A specialized conditioning protocol for honeybees that is designed for use within a complex agricultural ecosystem. This method ensures that the conditioned bees will be less likely to exhibit a conditioned response to uninfected plants, a false positive response that would render such a biological sensor unreliable for agricultural decision support. Also described is a superboosting training regime that allows training without the aid of expensive equipment and protocols for training in out in the field. Also described is a memory enhancing cocktail that aids in long term memory retention of a vapor signature. This allows the bees to be used in the field for longer durations and with fewer bees trained overall.

59 BASIC BIOLOGICAL SCIENCES↗

Autotune Calibrates Models to Building Use Data

Models of existing buildings are currently unreliable unless calibrated manually by a skilled professional. Autotune, as the name implies, automates this process by calibrating the model of an existing building to measured data, and is now available as open source software. This enables private businesses to incorporate Autotune into their products so that their customers can more effectively estimate cost savings of reduced energy consumption measures in existing buildings.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Learning from Trending, Precursor Analysis, and System Failures

Models of reliability growth relate current system unreliability to currently accumulated experience. But “experience” comes in different forms. Looking back after a major accident, one is sometimes able to identify previous events or measurable performance trends that were, in some sense, signaling the potential for that major accident: potential that could have been recognized and acted upon, but was not recognized until the accident occurred. This could be a previously unrecognized cause of accidents, or underestimation of the likelihood that a recognized potential cause would actually operate. Despite improvements in the state of practice of modeling of risk and reliability, operational experience still has a great deal to teach us, and work has been going on in several industries to try to do a better job of learning from experience before major accidents occur. It is not enough to say that we should review operating experience; there is too much “experience” for such general advice to be considered practical. The paper discusses the following: 1. The challenge of deciding what to focus on in analysis of operating experience. 2. Comparing what different models of learning and reliability growth imply about trending and precursor analysis.

97 MATHEMATICS AND COMPUTING↗