Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Training Time”

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 613 records · Page 34
All sourcesNASA NTRSDOE OSTI
NASA NTRS2022

Wallops Flight Facility Integration and Testing Customer Handbook

The Wallops Integration and Testing Handbook is a guide for planning operations at Wallops testing facilities. It provides a summary of the capabilities and technical specifications of available facilities. This handbook prescribes the information to be provided by the user that will enable the integration and testing team to effectively plan for and support the facility user’s project. Note that each facility plans for availability based on current workloads/ mission schedules. For example, sounding rockets may have between 20 and 35 missions in work at one time. Also note that special training, such as cleanroom certification, electrostatic discharge (ESD) or confined space training may be required for entrance to some facilities.

Adam Hammoudi↗
NASA NTRS2024

Prithvi-EO-2.0: A Versatile Multi-Temporal Foundation Model for Earth Observation Applications

This technical report presents Prithvi-EO-2.0, a new geospatial foundation model that offers significant improvements over its predecessor, Prithvi-EO-1.0. Trained on 4.2M global time series samples from NASA’s Harmonized Landsat and Sentinel-2 data archive at 30m resolution, the new 300M and 600M parameter models incorporate temporal and location embeddings for enhanced performance across various geospatial tasks. Through extensive benchmarking with GEOBench, the 600M version outperforms the previous Prithvi-EO model by 8% across a range of tasks. It also outperforms six other geospatial foundation models when benchmarked on remote sensing tasks from different domains and resolutions (i.e. from 0.1m to 15m). The results demonstrate the versatility of the model in both classical earth observation and high-resolution applications. Early involvement of end-users and subject matter experts (SMEs) are among the key factors that contributed to the project’s success. In particular, SME involvement allowed for constant feedback on model and dataset design, as well as successful customization for diverse SME-led applications in disaster response, land use and crop mapping, and ecosystem dynamics monitoring. Prithvi-EO-2.0 is available on Hugging Face and IBM terratorch, with additional resources on GitHub. The project exemplifies the Trusted Open Science approach embraced by all involved organizations.

Daniela Szwarcman↗
DOE OSTI2020

Theoretical Development of Controller Transfer applied to Dynamical Systems

Data-driven control strategies suffer from the necessity of training and its associated computational time and cost. This motivates the continuous increase of popularity of transfer learning based methods. In this paper, we have investigated transfer learning in the context of controller for dynamical systems. First, we have derived conditions under which controller can be transferred between two linear dynamical systems. Furthermore, we have introduced a novel algorithm to design the transferable and non-transferable control components for linear systems. Secondly, we have derived conditions under which a stabilizing controller can be transferred between two different nonlinear dynamical systems. Furthermore, we have used the notion of zero dynamics to transform the nonlinear dynamics into global normal form, to derive the transferability conditions on the controller. Finally, we have numerically evaluated the performance of our transfer learning based methods for two sets of example linear and nonlinear dynamical systems.

Chakraborty, Indrasis↗
DOE OSTI2022

Physics-driven learning of Wasserstein GAN for density reconstruction in dynamic tomography

Object density reconstruction from projections containing scattered radiation and noise is of critical importance in many applications. Existing scatter correction and density reconstruction methods may not provide the high accuracy needed in many applications and can break down in the presence of unmodeled or anomalous scatter and other experimental artifacts. Incorporating machine-learning models could prove beneficial for accurate density reconstruction, particularly in dynamic imaging, where the time evolution of the density fields could be captured by partial differential equations or by learning from hydrodynamics simulations. In this work, we demonstrate the ability of learned deep neural networks to perform artifact removal in noisy density reconstructions, where the noise is imperfectly characterized. Here, we use a Wasserstein generative adversarial network (WGAN), where the generator serves as a denoiser that removes artifacts in densities obtained from traditional reconstruction algorithms. We train the networks from large density time-series datasets, with noise simulated according to parametric random distributions that may mimic noise in experiments. The WGAN is trained with noisy density frames as generator inputs, to match the generator outputs to the distribution of clean densities (time series) from simulations. A supervised loss is also included in the training, which leads to an improved density restoration performance. In addition, we employ physics-based constraints such as mass conservation during the network training and application to further enable highly accurate density reconstructions. Our preliminary numerical results show that the models trained in our frameworks can remove significant portions of unknown noise in density time-series data.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Visual feedback and guided balance training in an immersive virtual reality environment for lower extremity rehabilitation

Balance training is essential for physical rehabilitation procedures, as it can improve functional mobility and enhance cognitive coordination. However, conventional balance training methods may have limitations in terms of motivation, real-time objective feedback, and personalization, which a virtual reality (VR) setup may provide a better alternative. In this work, we present an immersive VR training environment for lower extremity balance rehabilitation with real-time guidance and feedback. The VR training environment immerses the user in a 3D ice rink model where a virtual coach (agent) leads them through a series of balance poses, and the user controls a trainee avatar with their own movements. Here we developed two coaching styles: positive-reinforcement and autonomous-supportive, and two viewpoints of the trainee avatar: first-person and third-person. The proposed environment was evaluated in a user study with healthy, non-clinical participants (n = 16, 24.4 ± 5.7 years old, 9 females). Our results show that participants showed stronger performance in the positive-reinforcement style compared to the autonomous-supportive style. Additionally, in the third-person viewpoint, the participants exhibited more stability in the positive-reinforcement style compared to the autonomous-supportive style. For viewpoint, participants exhibited stronger performance in the first-person viewpoint compared to third-person in the autonomous-supportive style, while they were comparable in the positive-reinforcement style. We observed no significant effects on the foot height and number of mistakes. Furthermore, we report the analysis of user performance with balance training poses and subjective measures based on questionnaires to assess the user experience, usability, and task load. The proposed VR balance training could offer an interactive, adaptive, and engaging environment and open new potential research directions for lower extremity rehabilitation.

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

1D-Convolutional Neural Network Architecture for Generalized Time-series Segmentation

Time segmentation of experimental data is a common and often difficult task. Consequently, it is of interest to automate this type of segmentation to reduce manual inputs, which are labor intensive and less consistent. However, simple thresholding algorithms are often insufficiently robust due either to noise or inconsistent data. This paper proposes a simple 1D convolutional neural net (CNN) architecture as a generalized solution for typical time segmentation tasks. The layer architecture, training methods, and methods for simple customization will be described as well as the results of application to three separate data streams: facility condition segmentation, video highlight segmentation, and calorimeter time-series segmentation. In all three test cases the 1D-CNN performs better than tailored integral/derivative/thresholding algorithms across a range of signal-to-noise levels.

CNN↗
NASA NTRSDate not supplied

1D-Convolutional Neural Network Architecture for Generalized Time-series Segmentation

Time segmentation of experimental data is a common and often difficult task. Consequently, it is of interest to automate this type of segmentation to reduce manual inputs, which are labor intensive and less consistent. However, simple thresholding algorithms are often insufficiently robust due either to noise or inconsistent data. This paper proposes a simple 1D convolutional neural net (CNN) architecture as a generalized solution for typical time segmentation tasks. The layer architecture, training methods, and methods for simple customization will be described as well as the results of application to three separate data streams: facility condition segmentation, video highlight segmentation, and calorimeter time-series segmentation. In all three test cases the 1D-CNN performs better than tailored integral/derivative/thresholding algorithms across a range of signal-to-noise levels.

CNN↗
DOE OSTI2024

PV Generation and Load Forecasting for Adjuntas PR Community Microgrids

Existing frameworks to forecast time-series photovoltaic (PV) output power and consumer load for microgrid operations and controls assume a near-continuous availability of real-time input features from the field assets such as PV inverters, energy meters, and weather station. These incoming data points are used to periodically retrain models and update forecast snapshots over a moving horizon window, be it one hour-ahead, one-day ahead, or one-week ahead. However, such frameworks are not resilient to disruptions in data availability caused by losses in communications between the field sensors and data loggers. Hence, there is a need for programs that assume no availability of real-time microgrid asset data and still make reliable forecasts that can be used for decision-making. Such programs would be apt to function in extreme weather events such as hurricanes and would use lightweight recursive time-series models to independently forecast solar irradiance and ambient temperature, then compute PV power from those forecasts, as well as independently forecast consumer load. The codebase performs forecasting for the scenario of when the microgrid does not have a reliable access to forecasts or real-time observations of solar irradiance (I) and ambient temperature (AT) and load (Load) to be able to adequately forecast, in real-time, the PV power production or a business' load. In this case, using historical values of PV power and load, a univariate forecasting of generation and consumption are respectively made. The use-case in particular has two sub-scenarios: one, a normal 7-day ahead forecast where the unavailability of real-time data is assumed due to infrastructure issues such as loss of communication or sensor maintenance or service downtimes. Whereas a hurricane-caused unavailability of real-time data requires a second model trained specifically on historical hurricane days to be able to capture the extreme day behavior of generation in particular, and load if applicable. A gradient boosted regression tree comprises an ensemble of additive models that map between the input of historical values (be it irradiance, temperature, or load) and their corresponding output forecasts of a given horizon such that the individual learner predictions are summed up over the total number of such learners in the ensemble to produce an aggregate forecast. A weighting mechanism is applied to the training data in each iteration, where actual and forecast values are compared to penalize incorrect forecasts by increasing the weight and reducing it to reward correct forecasts. The code's benefits are that it: (a) accounts for a contingency where communication loss renders newly measured real-time data unavailable for model tuning and snapshot updates; (b) presents blind forecasting that recursively determines the next time-step value in a horizon using the forecast of the same attribute from a prior step; and (c) employs lightweight models that, once trained, can reliably generalize for different horizons, which make them suitable for enhancing the resilience of field microgrids prone to extreme events that encounter disruptions to data availability.

Sundararajan, Aditya [Oak Ridge National Laborator↗
NASA NTRSDate not supplied

1D-Convolutional Neural Network Architecture for Generalized Time-Segmentation Tasks

Time segmentation of experimental data is a common and often difficult task. Consequently, it is of interest to automate this type of segmentation to reduce manual inputs, which are labor intensive and less consistent. However, simple thresholding algorithms are often insufficiently robust due either to noise or inconsistent data. This paper proposes a simple 1D CNN architecture as a generalized solution for typical time segmentation tasks. The layer architecture, training methods, and methods for simple customization are described as well as the results of application to three separate arc jet data streams: facility condition segmentation, video highlight segmentation, and calorimeter time-series segmentation.

Machine learning↗
NASA NTRSDate not supplied

1d-Convolutional Neural Network Architecture for Generalized Time-Segmentation Tasks

Time segmentation of experimental data is a common and often difficult task. Consequently, it is of interest to automate this type of segmentation to reduce manual inputs, which are labor intensive and less consistent. However, simple thresholding algorithms are often insufficiently robust due either to noise or inconsistent data. This paper proposes a simple 1D CNN architecture as a generalized solution for typical time segmentation tasks. The layer architecture, training methods, and methods for simple customization are described as well as the results of application to three separate arc jet data streams: facility condition segmentation, video highlight segmentation, and calorimeter time-series segmentation.

Machine learning↗
NASA NTRSDate not supplied

1D-Convolutional Neural Network Architecture for Generalized Time-Segmentation Tasks

Time segmentation of experimental data is a common and often difficult task. Consequently, it is of interest to automate this type of segmentation to reduce manual inputs, which are labor intensive and less consistent. However, simple thresholding algorithms are often insufficiently robust due either to noise or inconsistent data. This paper proposes a simple 1D convolutional neural net (CNN) architecture as a generalized solution for typical time segmentation tasks. The layer architecture, training methods, and methods for simple customization are described as well as the results of application to three separate arc jet data streams: facility condition segmentation, video highlight segmentation, and calorimeter time-series segmentation.

Machine learning↗
NASA NTRS1995

Training for Aviation Decision Making: The Naturalistic Decision Making Perspective

This paper describes the implications of a naturalistic decision making (NDM) perspective for training air crews to make flight-related decisions. The implications are based on two types of analyses: (a) identification of distinctive features that serve as a basis for classifying a diverse set of decision events actually encountered by flight crews, and (b) performance strategies that distinguished more from less effective crews flying full-mission simulators, as well as performance analyses from NTSB accident investigations. Six training recommendations are offered: (1) Because of the diversity of decision situations, crews need to be aware that different strategies may be appropriate for different problems; (2) Given that situation assessment is essential to making a good decision, it is important to train specific content knowledge needed to recognize critical conditions, to assess risks and available time, and to develop strategies to verify or diagnose the problem; (3) Tendencies to oversimplify problems may be overcome by training to evaluate options in terms of goals, constraints, consequences, and prevailing conditions; (4) In order to provide the time to gather information and consider options, it is essential to manage the situation, which includes managing crew workload, prioritizing tasks, contingency planning, buying time (e.g., requesting holding or vectors), and using low workload periods to prepare for high workload; (5) Evaluating resource requirements ("What do I need?") and capabilities ("'What do I have?" ) are essential to making good decisions. Using resources to meet requirements may involve the cabin crew, ATC, dispatchers, and maintenance personnel; (6) Given that decisions must often be made under high risk, time pressure, and workload, train under realistic flight conditions to promote the development of robust decision skills.

Orasanu, Judith↗
DOE OSTI2025

SigTime: Learning and Visually Explaining Time Series Signatures

Understanding and distinguishing temporal patterns in time series data is essential for scientific discovery and decision-making. For example, in biomedical research, uncovering meaningful patterns in physiological signals can improve diagnosis, risk assessment, and patient outcomes. However, existing methods for time series pattern discovery face major challenges, including high computational complexity, limited interpretability, and difficulty in capturing meaningful temporal structures. Here, to address these gaps, we introduce a novel learning framework that jointly trains two Transformer models using complementary time series representations: shapelet-based representations to capture localized temporal structures and traditional feature engineering to encode statistical properties. The learned shapelets serve as interpretable signatures that differentiate time series across classification labels. Additionally, we develop a visual analytics system—SigTime—with coordinated views to facilitate exploration of time series signatures from multiple perspectives, aiding in useful insights generation. We quantitatively evaluate our learning framework on eight publicly available datasets and one proprietary clinical dataset. Additionally, we demonstrate the effectiveness of our system through two usage scenarios along with the domain experts: one involving public ECG data and the other focused on preterm labor analysis.

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

Physics and Modeling of Quasi-Steady and Forced Shock Trains

The dual-mode scramjet is a propulsive concept for contemporary hypersonic vehicles promising flight cruise speeds upwards of Mach 5. Despite decades of development and a recent resurgence in interest, such engines are not routinely deployed for commercial or military purposes. Instead, they have been confined to the realm of ground-test research and limited one-off flight test missions. One of the key stumbling blocks on the critical path to deployment of a dual-mode scramjet is the optimal design and stable operation of the isolator, a typically constant area duct that separates the scramjet inlet from the combustor. During ramjet mode at Mach numbers from 3 to 5 and during the transition to scramjet mode at higher Mach numbers, the isolator contains the pseudo-shock. Necessary to meet the back-pressure in the combustor, the pseudo-shock is typically divided into a shock dominated shock train region followed by a turbulence driven mixing region. Fluctuations in combustor pressure due to vehicle maneuvering, throttling, or combustion instabilities drive the shock train to shift streamwise within the isolator. In response to sharp pressure fluctuations, the shock train is unable to stabilize within the isolator, and the scramjet unstarts by ejecting the shock train out the inlet. This work addresses three key gaps in understanding of the isolator pseudo-shock. The phenomenon is studied in three experimental facilities to investigate the quasi-steady and unsteady fluid dynamics of the shock train in fueled and un-fueled test conditions. The strong link between boundary layer morphology and shock train structure is studied in detail and used as a tool to evaluate the state-of-the-art definitions and models of the pseudo-shock. Identifying the limitations of these models, a new process driven flux-conserved pseudo-shock model is proposed. This provides a computationally efficient model grounded in the fluid dynamic mechanisms of the pseudo-shock with better predictive capability than the current state-of-the-art across all facilities tested. The unsteady behavior of the shock train is further investigated to identify and understand two distinct forced dynamic modes: mechanically induced, low-rate forcing and high-rate combustion driven unsteadiness that sheds upstream propagating shocks. The proposed process driven model is found to be effective in the the limiting case of high-rate combustion driven unsteadiness by evaluating the model in a shock stationary reference frame. Further analysis identifies the approximately rigid chain motion of the shock train structure to mechanically induced forcing. The time delay between the onset of forced dynamics is found to saturate to a constant value and maximum shock train speed along the isolator with respect to the perturbation rate of change. The result of this work provides dual-mode scramjet designers with improved tools and physical understanding to develop more efficient, robust, and capable propulsion systems for hypersonic vehicles.

JAXA↗
NASA NTRS2017

A MODIS-Based Automated Flood Monitoring System for Southeast Asia

Flood disasters in Southeast Asia result in significant loss of life and economic damage. Remote sensing information systems designed to spatially and temporally monitor floods can help governments and international agencies formulate effective disaster response strategies during a flood and ultimately alleviate impacts to population, infrastructure, and agriculture. Recent destructive flood events in the Lower Mekong River Basin occurred in 2000, 2011, 16 2013, and 2016 (http://ffw.mrcmekong.org/historical_rec.htm, April 24, 2017). The large spatial distribution of flooded areas and lack of proper gauge data in the region makes accurate monitoring and assessment of impacts of floods difficult. Here, we discuss the utility of applying satellite-based Earth observations for improving flood inundation monitoring over the flood-prone Lower Mekong River Basin. We present a methodology for determining near real-time surface water extent associated with current and historic flood events by training surface water classifiers from 8-day, 250-meter Moderate-resolution Imaging Spectroradiometer (MODIS) data spanning the length of the MODIS satellite record. The Normalized Difference Vegetation Index (NDVI) signature of permanent water bodies (MOD44W; Carroll et al., 2009) is used to train surface water classifiers which are applied to a time period of interest. From this, an operational nowcast flood detection component is produced using twice daily imagery acquired at 3-hour latency which performs image compositing routines to minimize cloud cover. Case studies and accuracy assessments against radar-based observations for historic flood events are presented. The customizable system has been transferred to regional organizations and near real-time derived surface water products are made available through a web interface platform. Results highlight the potential of near real-time observation and impact assessment systems to serve as effective decision support tools for governments, international agencies, and disaster responders.

Mekong River↗
NASA NTRS2016

Case Study: Managing Undergraduate Interns in the Context of Multidisciplinary Projects

Internship opportunities can be immensely beneficial to both the students who pursue them and the labs that employ them. However, optimal results for all stakeholders can be uncertain. Several factors compound this issue. In this case study, the highly multidisciplinary nature of the BioEngineering Instrumentation Group at NASA Ames Research Center has made productively integrating short-term summer interns challenging. In general, taking on more undergraduate interns during a given time period can lead to diminishing returns, particularly if those interns are primarily short-time interns with few long-term senior students present.Introducing a level of industry-style project management processes can help. On the one hand, extreme formality can reduce the enthusiasm that students can bring; also, time constraints on internships make extensive training in formal process an investment with comparatively little return. On the other hand, lack of process can result in many forms of wastage. A moderate, lightweight level of process i.e., simple processes with high payoff to time invested ratios, including processes to effectively handle multidisciplinary issues can yield both a far more efficient lab, and a richer educational experience for the interns. In this case study, we implemented simple processes such as brief weekly stand-up meetings, a standard action item bug tracking process, and deadline-based scheduling with prudent reserves and critical path tracking.To evaluate these steps, several parameters were considered: the number of projects the lab could successfully pursue, the number of interns that could be successfully integrated onto each project, diversity of student intern educational backgrounds, student satisfaction, student understanding of expectations, and student preparation for industry-standard work. Generally, outcomes were enhanced across the board, even with imperfectly implemented processes. Lightweight process implementation is shown to be tremendously effective in a multidisciplinary internship context.

Case Study↗
NASA NTRS1991

Neural Architectures for Control

The cerebellar model articulated controller (CMAC) neural architectures are shown to be viable for the purposes of real-time learning and control. Software tools for the exploration of CMAC performance are developed for three hardware platforms, the MacIntosh, the IBM PC, and the SUN workstation. All algorithm development was done using the C programming language. These software tools were then used to implement an adaptive critic neuro-control design that learns in real-time how to back up a trailer truck. The truck backer-upper experiment is a standard performance measure in the neural network literature, but previously the training of the controllers was done off-line. With the CMAC neural architectures, it was possible to train the neuro-controllers on-line in real-time on a MS-DOS PC 386. CMAC neural architectures are also used in conjunction with a hierarchical planning approach to find collision-free paths over 2-D analog valued obstacle fields. The method constructs a coarse resolution version of the original problem and then finds the corresponding coarse optimal path using multipass dynamic programming. CMAC artificial neural architectures are used to estimate the analog transition costs that dynamic programming requires. The CMAC architectures are trained in real-time for each obstacle field presented. The coarse optimal path is then used as a baseline for the construction of a fine scale optimal path through the original obstacle array. These results are a very good indication of the potential power of the neural architectures in control design. In order to reach as wide an audience as possible, we have run a seminar on neuro-control that has met once per week since 20 May 1991. This seminar has thoroughly discussed the CMAC architecture, relevant portions of classical control, back propagation through time, and adaptive critic designs.

Peterson, James K.↗
NASA NTRS1991

Design of thrust vectoring exhaust nozzles for real-time applications using neural networks

Thrust vectoring continues to be an important issue in military aircraft system designs. A recently developed concept of vectoring aircraft thrust makes use of flexible exhaust nozzles. Subtle modifications in the nozzle wall contours produce a non-uniform flow field containing a complex pattern of shock and expansion waves. The end result, due to the asymmetric velocity and pressure distributions, is vectored thrust. Specification of the nozzle contours required for a desired thrust vector angle (an inverse design problem) has been achieved with genetic algorithms. This approach is computationally intensive and prevents the nozzles from being designed in real-time, which is necessary for an operational aircraft system. An investigation was conducted into using genetic algorithms to train a neural network in an attempt to obtain, in real-time, two-dimensional nozzle contours. Results show that genetic algorithm trained neural networks provide a viable, real-time alternative for designing thrust vectoring nozzles contours. Thrust vector angles up to 20 deg were obtained within an average error of 0.0914 deg. The error surfaces encountered were highly degenerate and thus the robustness of genetic algorithms was well suited for minimizing global errors.

Prasanth, Ravi K.↗