Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “online training”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
NASA NTRS2009

Methodologies for Adaptive Flight Envelope Estimation and Protection

This paper reports the latest development of several techniques for adaptive flight envelope estimation and protection system for aircraft under damage upset conditions. Through the integration of advanced fault detection algorithms, real-time system identification of the damage/faulted aircraft and flight envelop estimation, real-time decision support can be executed autonomously for improving damage tolerance and flight recoverability. Particularly, a bank of adaptive nonlinear fault detection and isolation estimators were developed for flight control actuator faults; a real-time system identification method was developed for assessing the dynamics and performance limitation of impaired aircraft; online learning neural networks were used to approximate selected aircraft dynamics which were then inverted to estimate command margins. As off-line training of network weights is not required, the method has the advantage of adapting to varying flight conditions and different vehicle configurations. The key benefit of the envelope estimation and protection system is that it allows the aircraft to fly close to its limit boundary by constantly updating the controller command limits during flight. The developed techniques were demonstrated on NASA s Generic Transport Model (GTM) simulation environments with simulated actuator faults. Simulation results and remarks on future work are presented.

Tang, Liang↗
DOE OSTI2024

Online MCMC Thinning with Kernelized Stein Discrepancy

A fundamental challenge in Bayesian inference is efficient representation of a target distribution. Many nonparametric approaches do so by sampling a large number of points using variants of Markov chain Monte Carlo (MCMC). Here, we propose an MCMC variant that retains only those posterior samples which exceed a kernelized Stein discrepancy (KSD) threshold, which we call KSD thinning. We establish the convergence and complexity trade-offs for several settings of KSD thinning as a function of the KSD threshold parameter, sample size, and other problem parameters. We provide experimental comparisons against other online nonparametric Bayesian methods that generate low-complexity posterior representations. We observe superior consistency/complexity trade-offs across a range of settings including MCMC sampling on two Bayesian inference problems from the biological sciences, and 10 × inference speedup and storage reduction for Bayesian neural networks with no loss of accuracy and no increase in training time. Our code is available at https://github.com/colehawkins/KSD-Thinning.

Bayesian inference↗
NASA NTRS1993

The application of automated operations at the Institutional Processing Center

The JPL Institutional and Mission Computing Division, Communications, Computing and Network Services Section, with its mission contractor, OAO Corporation, have for some time been applying automation to the operation of JPL's Information Processing Center (IPC). Automation does not come in one easy to use package. Automation for a data processing center is made up of many different software and hardware products supported by trained personnel. The IPC automation effort formally began with console automation, and has since spiraled out to include production scheduling, data entry, report distribution, online reporting, failure reporting and resolution, documentation, library storage, and operator and user education, while requiring the interaction of multi-vendor and locally developed software. To begin the process, automation goals are determined. Then a team including operations personnel is formed to research and evaluate available options. By acquiring knowledge of current products and those in development, taking an active role in industry organizations, and learning of other data center's experiences, a forecast can be developed as to what direction technology is moving. With IPC management's approval, an implementation plan is developed and resources identified to test or implement new systems. As an example, IPC's new automated data entry system was researched by Data Entry, Production Control, and Advance Planning personnel. A proposal was then submitted to management for review. A determination to implement the new system was made and elements/personnel involved with the initial planning performed the implementation. The final steps of the implementation were educating data entry personnel in the areas effected and procedural changes necessary to the successful operation of the new system.

Barr, Thomas H.↗
DOE OSTI2021

Accurate prediction of mega-electron-volt electron beam properties from UED using machine learning

To harness the full potential of the ultrafast electron diffraction (UED) and microscopy (UEM), we must know accurately the electron beam properties, such as emittance, energy spread, spatial-pointing jitter, and shot-to-shot energy fluctuation. Owing to the inherent fluctuations in UED/UEM instruments, obtaining such detailed knowledge requires real-time characterization of the beam properties for each electron bunch. While diagnostics of these properties exist, they are often invasive, and many of them cannot operate at a high repetition rate. Here, we present a technique to overcome such limitations. Employing a machine learning (ML) strategy, we can accurately predict electron beam properties for every shot using only parameters that are easily recorded at high repetition rate by the detector while the experiments are ongoing, by training a model on a small set of fully diagnosed bunches. Applying ML as real-time noninvasive diagnostics could enable some new capabilities, e.g., online optimization of the long-term stability and fine single-shot quality of the electron beam, filtering the events and making online corrections of the data for time-resolved UED, otherwise impossible. This opens the possibility of fully realizing the potential of high repetition rate UED and UEM for life science and condensed matter physics applications.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
NASA NTRS2011

Feature Acquisition with Imbalanced Training Data

This work considers cost-sensitive feature acquisition that attempts to classify a candidate datapoint from incomplete information. In this task, an agent acquires features of the datapoint using one or more costly diagnostic tests, and eventually ascribes a classification label. A cost function describes both the penalties for feature acquisition, as well as misclassification errors. A common solution is a Cost Sensitive Decision Tree (CSDT), a branching sequence of tests with features acquired at interior decision points and class assignment at the leaves. CSDT's can incorporate a wide range of diagnostic tests and can reflect arbitrary cost structures. They are particularly useful for online applications due to their low computational overhead. In this innovation, CSDT's are applied to cost-sensitive feature acquisition where the goal is to recognize very rare or unique phenomena in real time. Example applications from this domain include four areas. In stream processing, one seeks unique events in a real time data stream that is too large to store. In fault protection, a system must adapt quickly to react to anticipated errors by triggering repair activities or follow- up diagnostics. With real-time sensor networks, one seeks to classify unique, new events as they occur. With observational sciences, a new generation of instrumentation seeks unique events through online analysis of large observational datasets. This work presents a solution based on transfer learning principles that permits principled CSDT learning while exploiting any prior knowledge of the designer to correct both between-class and withinclass imbalance. Training examples are adaptively reweighted based on a decomposition of the data attributes. The result is a new, nonparametric representation that matches the anticipated attribute distribution for the target events.

Thompson, David R.↗
DOE OSTI2020

Machine Learning-Based PV Reserve Determination Strategy for Frequency Control on the WECC System

Frequency control from photovoltaic (PV) power plants has great potential to address the frequency response challenge of the power system with high penetrations of renewable generation. Using model-based approaches to determine the optimal PV headroom reserve, however, requires significant online computation and is intractable for an interconnection level system. This paper proposes a machine learning based strategy, that is suitable for real-time operation, to determine the optimal PV reserve for frequency control. The proposed machine learning algorithm is trained and tested on 1,987 offline simulations of a 60% renewable penetration Western Electricity Coordinating Council (WECC) system. Furthermore, the proposed reserve determination strategy is applied on a realistic 1-day operation profile of the WECC system and demonstrates a savings of more than 40% PV headroom compared to a conservative approach. It is evident that the proposed strategy can efficiently and effectively determine the optimal PV frequency control reserve for realistic interconnection systems.

frequency control↗
DOE OSTI2022

Online Power System Event Detection via Bidirectional Generative Adversarial Networks

Accurate and speedy detection of power system events is critical to enhancing the reliability and resiliency of power systems. Although supervised deep learning algorithms show great promise in identifying power system events, they require a large volume of high-quality event labels for training. This paper develops a bidirectional anomaly generative adversarial network (GAN)-based algorithm to detect power system events using streaming PMU data, which does not rely on a huge amount of event labels. By introducing conditional entropy constraint in the objective function of GAN and graph signal processing-based PMU sorting technique, our proposed algorithm significantly outperforms state-of-the-art event detection algorithms in terms of accuracy. To facilitate the adoption of the proposed algorithm, a prototype online platform is also developed using Apache Hadoop, Kafka, and Spark to enable real-time event detection. Here, the accuracy and computational efficiency of the proposed algorithm are validated using a large-scale real-world PMU dataset from the Eastern Interconnection of the United States.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2024

Online Learning of Effective Turbine Wind Speed in Wind Farms

To develop better wind farm controllers that can meet more complex objectives, methods of modeling the wind turbine wakes at low computational expense are needed. Gaussian process (GP) regression offers a computationally inexpensive framework for learning complex functions from noisy measurements with very few datapoints. In this work, an online learning approach is presented to learn the rotor-averaged wind velocity at downstream wind turbines with GPs, using the available datastream of wind field measurements and wind turbine control set-points. This framework can readily be integrated into model-based controls methods because the model a) is updated online at low computational expense, b) assumes a mathematically favorable Gaussian form, and c) explicitly quantifies the stochastic nature of the wake field so that the trade-off between exploration and exploitation, and the uncertainty in the prediction, can be utilized. We show that a GP-learned model can match true values with errors within 0.5% on average, with as few as 5 training data points.

Gaussian process↗
NASA NTRS2007

A-Train Data Depot: Integrating and Exploring Data Along the A-Train Tracks

The immense potential for new science findings as a result of inter-instrument data analysis has led to the development of a new data portal at GSFC: the A-train Data Depot. The power and utility of this new service to the general public is amplified immensely when the archived data are used in conjunction with online data analysis services like Giovanni. This presentation details some of the challenges of data usage from multiple distinct missions and how the tool sets we have developed can help to overcome these challenges, considerably cut down on analysis overhead and promote science exploration in an otherwise very challenging arena.

Leptoukh, G.↗
NASA NTRSDate not supplied

An Online Reinforcement Learning Controller Design For Mars Ascent Vehicle

This paper presents a neural network (NN) approximator-based online reinforcement learning (ORL) controller design for Mars Ascent Vehicle (MAV) under parametric variation and significant external disturbances. The ORL controller, which does not require any offline training, involves two NNs where an action NN produces optimal short-term control performance while a critic NN evaluates the performance of the action NN using an approximated cost function. The simulation example with comparisons against baseline Proportional-Integral-Derivative (PID) and gain scheduled pole-placement PID (GS-PP-PID) controllers show the proposed controller’s effectiveness and robustness under parametric variation and high external disturbances.

Han Woong Bae↗
DOE OSTI2023

Enabling scientific machine learning in MOOSE using Libtorch

A neural-network-based machine learning interface has been developed for the Multiphysics Object-Oriented Simulation Environment (MOOSE). The interface relies on Libtorch, the C++ front-end of PyTorch, and enables an online interaction between modern machine learning algorithms and all the existing simulation, modeling, and analysis processes available in MOOSE. New capabilities in MOOSE include the native generation and training of artificial neural networks together with options to load pretrained neural networks in TorchScript format. Furthermore, the MOOSE stochastic tools module (MOOSE-STM) has been enhanced with neural network-based surrogate and reduced-order model generation options for efficient stochastic analyses. Lastly, a reinforcement learning capability has been added to MOOSE-STM for the interactive control and optimization of complex multiphysics problems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Probabilistic error estimation for non-intrusive reduced models learned from data of systems governed by linear parabolic partial differential equations

This work derives a residual-based a posteriori error estimator for reduced models learned with non-intrusive model reduction from data of high-dimensional systems governed by linear parabolic partial differential equations with control inputs. It is shown that quantities that are necessary for the error estimator can be either obtained exactly as the solutions of least-squares problems in a non-intrusive way from data such as initial conditions, control inputs, and high-dimensional solution trajectories or bounded in a probabilistic sense. Here, the computational procedure follows an offline/online decomposition. In the offline (training) phase, the high-dimensional system is judiciously solved in a black-box fashion to generate data and to set up the error estimator. In the online phase, the estimator is used to bound the error of the reduced-model predictions for new initial conditions and new control inputs without recourse to the high-dimensional system. Numerical results demonstrate the workflow of the proposed approach from data to reduced models to certified predictions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Neural network approaches for parameterized optimal control

Here, we consider numerical approaches for deterministic, finite-dimensional optimal control problems whose dynamics depend on unknown or uncertain parameters. We seek to amortize the solution over a set of relevant parameters in an offline stage to enable rapid decision-making and be able to react to changes in the parameter in the online stage. To tackle the curse of dimensionality arising when the state and/or parameter are high-dimensional, we represent the policy using neural networks. We compare two training paradigms: First, our model-based approach leverages the dynamics and definition of the objective function to learn the value function of the parameterized optimal control problem and obtain the policy using a feedback form. Second, we use actor-critic reinforcement learning to approximate the policy in a data-driven way. Using an example involving a two-dimensional convection-diffusion equation, which features high-dimensional state and parameter spaces, we investigate the accuracy and efficiency of both training paradigms. While both paradigms lead to a reasonable approximation of the policy, the model-based approach is more accurate and considerably reduces the number of PDE solves.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1996

Software For Multivariate Bayesian Classification

PHD general-purpose classifier computer program. Uses Bayesian methods to classify vectors of real numbers, based on combination of statistical techniques that include multivariate density estimation, Parzen density kernels, and EM (Expectation Maximization) algorithm. By means of simple graphical interface, user trains classifier to recognize two or more classes of data and then use it to identify new data. Written in ANSI C for Unix systems and optimized for online classification applications. Embedded in another program, or runs by itself using simple graphical-user-interface. Online help files makes program easy to use.

Saul, Ronald↗
NASA NTRS2018

Online Self-Supervised Long-Range Scene Segmentation for MAVs

Recently, there have been numerous advances in the development of payload and power constrained lightweight Micro Aerial Vehicles (MAVs). As these robots aspire for high-speed autonomous flights in complex dynamic environments, robust scene understanding at long-range becomes critical. The problem is heavily characterized by either the limitations imposed by sensor capabilities for geometry-based methods, or the need for large-amounts of manually annotated training data required by data-driven methods. This motivates the need to build systems that have the capability to alleviate these problems by exploiting the complimentary strengths of both geometry and data-driven methods. In this paper, we take a step in this direction and propose a generic framework for adaptive scene segmentation using self-supervised online learning. We present this in the context of vision-based autonomous MAV flight, and demonstrate the efficacy of our proposed system through extensive experiments on benchmark datasets and real world field tests.

Matthies, Larry↗
DOE OSTI2020

Evaluating the Adaptability of Reinforcement Learning Based HVAC Control for Residential Houses

Intelligent Heating, Ventilation, and Air Conditioning (HVAC) control using deep reinforcement learning (DRL) has recently gained a lot of attention due to its ability to optimally control the complex behavior of the HVAC system. However, more exploration is needed on understanding the adaptability challenges that the DRL agent could face during the deployment phase. Using online learning for such applications is not realistic due to the long learning period and likely poor comfort control during the learning process. Alternatively, DRL can be pre-trained using a building model prior to deployment. However, developing an accurate building model for every house and deploying a pre-trained DRL model for HVAC control would not be cost-effective. In this study, we focus on evaluating the ability of DRL-based HVAC control to provide cost savings when pre-trained on one building model and deployed on different house models with varying user comforts. We observed around 30% of cost reduction by pre-trained model over baseline when validated in a simulation environment and achieved up to 21% cost reduction when deployed in the real house. This finding provides experimental evidence that the pre-trained DRL has the potential to adapt to different house environments and comfort settings.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
NASA NTRS2008

A-Train Data Depot (ATDD)

ATDD is a web based tool which provides collocated data and display products for a number of A-train instruments Cloudsat, Calipso, OMI, AIRS, MODIS, MLS, POLDER-3, and ECWMF model data. Products provided include Clouds, Aerosols, Water Vapor, Temperatures and trace gases. All input data is online and in HDF4, HDF5 format. Display products include curtain images, horizontal strips, line plot overlays, and GE kmz files. Sample products are shown for two type of events. Hurricane event, Norbert, Oct 8, 2008 and a dust storm event over the Arabian Sea, Nov 13-14, 2008.

Smith, Peter M.↗
NASA NTRS2013

Sensory Supplementation to Enhance Adaptation Following G-transitions and Traumatic Brain Injury

Sensory supplementation can be incorporated as online feedback for improving spatial orientation awareness for manual control tasks (e.g. TSAS, Shuttle ZAG study). Preliminary data with vestibular patients and TBI military population is promising for rehabilitation training. Recommend that sensory supplementation be incorporated as a training component in an integrated countermeasure approach.

Wood, Scott↗