Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Feature Engineering”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
NASA NTRS1982

Energy efficient engine component development and integration program

The objective of the Energy Efficient Engine Component Development and Integration program is to develop, evaluate, and demonstrate the technology for achieving lower installed fuel consumption and lower operating costs in future commercial turbofan engines. Minimum goals have been set for a 12 percent reduction in thrust specific fuel consumption (TSFC), 5 percent reduction in direct operating cost (DOC), and 50 percent reduction in performance degradation for the Energy Efficient Engine (flight propulsion system) relative to the JT9D-7A reference engine. The Energy Efficienct Engine features a twin spool, direct drive, mixed flow exhaust configuration, utilizing an integrated engine nacelle structure. A short, stiff, high rotor and a single stage high pressure turbine are among the major enhancements in providing for both performance retention and major reductions in maintenance and direct operating costs. Improved clearance control in the high pressure compressor and turbines, and advanced single crystal materials in turbine blades and vanes are among the major features providing performance improvement. Highlights of work accomplished and programs modifications and deletions are presented.

Source record↗
NASA NTRSDate not supplied

A materials-informatics based study of solid electrolytes and protective coatings for Li batteries

All-solid-state batteries with Li metal anode can address the safety issues surrounding traditional Li-ion batteries as well as the demand for higher energy densities. However, the development of solid electrolytes and protective coatings simultaneously possessing high ionic conductivity and wide electrochemical stability has proven to be a challenge. Here, we present a data-driven approach to explore the Li compound space for promising solid electrolytes and coatings. This is accomplished through the generation of a large database of battery-related materials properties of Li compounds by computing Li+ migration barriers using bond-valence-based pair potentials, and stability windows using density functional theory energies. Using this database, we implement machine learning models that can accurately predict migration barriers and electrochemical stability windows for any new Li compound. Through feature engineering, we ensure that our models are both accurate and interpretable. We perform feature importance analysis on our models to highlight materials properties that can be tuned for future design of coatings/electrolytes. Our database and informatics approach provide a valuable tool for the rapid discovery of new solid-state battery chemistries.

Solid state batteries↗
DOE OSTI2026

Leveraging large language models to address data scarcity in machine learning for graphene synthesis

Machine learning in experimental materials science faces significant challenges due to the scarcity of data, which are costly and time-consuming to generate, particularly when relying on in-house experiments. Literature data mining offers a potential solution but introduces issues like mixed data quality, inconsistent formats, and non-uniform reporting of synthesis parameters, resulting in partially missing and heterogeneous features across the dataset. Here, we propose data imputation and feature engineering methods that employ pre-trained large language models (LLMs) to enhance machine learning performance on scarce, heterogeneous datasets, demonstrated on graphene CVD synthesis data and the ML-HydPARK hydrogen storage dataset. GPT models perform data imputation via tailored prompting and semantic normalization of inconsistently reported features through embeddings, for example, to harmonize the complex nomenclature of CVD substrates. Beyond yielding more diverse and richer feature representations than traditional methods such as K-nearest neighbors (KNN) and Multivariate Imputation by Chained Equations (MICE), LLM-based data imputation is evaluated against dataset characteristics and prompting strategies. We vary the level of autonomy granted to the LLM, from generic prompting that leverages pre-trained knowledge for autonomous data generation to data-informed prompting that constrains outputs using target-specific information, and demonstrate which level of autonomy yields superior imputation performance across datasets and feature types. The proposed data engineering methods markedly improve downstream performance; for example, in graphene layer number classification using a support vector machine (SVM), binary accuracy increases from 39% to 65% and ternary accuracy from 52% to 72%. Fine-tuning experiments on both datasets show that combining our proposed LLM-based data imputation and feature encoding methods with numerical machine learning predictors outperforms standalone fine-tuned LLM predictors in data-scarce settings. The proposed strategies emphasize data enhancement techniques rather than refining learning architectures or regularizing loss functions, offering a broadly applicable framework for improving machine learning performance on scarce, inhomogeneous datasets.

Chemical vapor deposition↗
NASA NTRS2006

The Nutating Engine-Prototype Engine Progress Report and Test Results

A prototype of a new, internal combustion (IC) engine concept has been completed. The Nutating Engine features an internal disk nutating (wobbling) on a Z-shaped power shaft. The engine is exceedingly compact, and several times more power dense than any conventional (reciprocating or rotary) IC engine. This paper discusses lessons learned during the prototype engine's development and provides details of its construction. In addition, results of the initial performance tests of the various components, as well as the complete engine, are summarized.

Meitner, Peter L.↗
NASA NTRS1988

Liquid rocket booster study. Volume 2, book 4, appendices 6-8: Reports of Rocketdyne, Pratt and Whitney, and TRW

For the pressure fed engines, detailed trade studies were conducted defining engine features such as thrust vector control methods, thrust chamber construction, etc. This was followed by engine design layouts and booster propulsion configuration layouts. For the pump fed engines parametric performance and weight data was generated for both O2/H2 and O2/RP-1 engines. Subsequent studies resulted in the selection of both LOX/RP-1 and O2/H2 propellants for the pump fed engines. More detailed analysis of the selected LOX/RP-1 and O2/H2 engines was conducted during the final phase of the study.

Source record↗
DOE OSTI2024

Feature selection with distance correlation

Choosing which properties of the data to use as input to multivariate decision algorithms—also known as feature selection—is an important step in solving any problem with machine learning. While there is a clear trend towards training sophisticated deep networks on large numbers of relatively unprocessed inputs (so-called automated feature engineering), for many tasks in physics, sets of theoretically well-motivated and well-understood features already exist. Working with such features can bring many benefits, including greater interpretability, reduced training and run time, and enhanced stability and robustness. We develop a new feature selection method based on distance correlation, and demonstrate its effectiveness on the tasks of boosted top- and W -tagging. Using our method to select features from a set of over 7,000 energy flow polynomials, we show that we can match the performance of much deeper architectures, by using only ten features and two orders-of-magnitude fewer model parameters. Published by the American Physical Society 2024

Astronomy & Astrophysics↗
DOE OSTI2025

Techno-economic analysis of sugarcane bagasse and straw conversion into cellulosic ethanol via consolidated bioprocessing

Cellulosic biofuels offer a sustainable alternative to fossil fuels and a means to mitigate climate change. Consolidated bioprocessing (CBP) featuring engineered thermophilic bacteria, combined with mechanical disruption during fermentation (cotreatment), has potential to lower production costs compared to featuring thermochemical pretreatment and added cellulase. A techno-economic analysis was conducted (230 million L ethanol/year) from sugarcane bagasse and straw at stand-alone facilities generating electricity from residues. Three scenarios were evaluated: Conventional, featuring hydrothermal pretreatment, fungal cellulase, and yeast fermentation (current commercial standard); Mid-term CBP, relying on bagasse solubilization without pretreatment or cotreatment; and Mature CBP, incorporating cotreatment but no pretreatment. Results for these scenarios in this order were: fixed capital investment (CapEx) $\$$589M, $\$$658M, and $\$$472M; net annual revenue (EBITDA) $\$$56M, $\$$96M, and $\$$94M; and minimum ethanol selling price 0.73, 0.61, and 0.48 US$\$$/L. Payback periods were 10.5, 6.9, and 5.1 years, while all scenarios showed <5 years at European prices for scales >100M L/year. Sensitivity and risk analysis highlighted ethanol price as the most critical variable. It is notable that Mid-term CBP had shorter payback times and better overall economic feasibility compared to Conventional. Our results underscore opportunities for research-driven innovation on low-cost cellulosic ethanol technologies in Brazil and elsewhere.

Consolidated bioprocessing↗
NASA NTRS1973

Design study of RL10 derivatives. Volume 2: Engine design characteristics

The design characteristics of the RL-10 rocket engine are discussed. The results from critical elements evaluation, baseline engine design, parametric and special study tasks are presented. Critical element evaluation established the feasibility of various engine features such as tank head idle, pumped idle, autogenous tank pressurization, and two-phase pumping. Three baseline engines, derived from the RL-10 were conceptually designed. Parametric life and performance data were generated. Special studies were conducted to establish the impact on the engine design of environment, safety, interchangeability, and maintenance.

Adams, A.↗
DOE OSTI2023

The dynamics, mixing, and thermonuclear burn of compressed foams with varied gas fills

Inertial confinement fusion (ICF) implosions involve highly coupled physics and complex hydrodynamics that are challenging to model computationally. Due to the sensitivity of such implosions to small features, detailed simulations require accurate accounting of the geometry and dimensionality of the initial conditions, including capsule defects and engineering features such as fill tubes used to insert gas into the capsule, yet this is computationally prohibitive. It is therefore difficult to evaluate whether discrepancies between the simulation and experiment arise from inadequate fidelity to the capsule geometry and drive conditions, uncertainties in physical data used by simulations, or inadequate physics. We present results from detailed high-resolution three-dimensional simulations of ICF implosions performed as part of the MARBLE campaign on the National Ignition Facility [Albright et al., Phys. Plasmas 29, 022702 (2022)]. These experiments are foam-filled separated-reactant experiments, where deuterons reside in the foam and tritons reside in the capsule gas fill and deuterium–tritium (DT) fusion reactions only occur in the presence of mixing between these materials. Material mixing in these experiments is primarily seeded by shock interaction with the complex geometry of the foam and gas fill, which induces the Richtmyer–Meshkov instability. We compare results for experiments with two different gas fills (ArT and HT), which lead to significant differences in the hydrodynamic and thermodynamic developments of the materials in the implosion. Our simulation results show generally good agreement with experiments and demonstrate a substantial impact of hydrodynamic flows on measured ion temperatures. The results suggest that viscosity, which was not included in our simulations, is the most important unmodeled physics and qualitatively explains the few discrepancies between the simulation and experiment. The results also suggest that the hydrodynamic treatment of shocks is inadequate to predict the heating and yield produced during shock flash, when the shock converges at the center of the implosion. Alternatively, underestimation of the level of radiative preheat from the shock front could explain many of the differences between the experiment and simulation. Nevertheless, simulations are able to reproduce many experimental observables within the level of experimental reproducibility, including most yields, time-resolved X-ray self-emission images, and an increase in burn-weighted ion temperature and neutron down-scattered ratio in the line of sight that includes a jet seeded by the glue spot that joins capsule hemispheres.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2021

Real-Time Health Monitoring for Gas Turbine Components Using Online Learning and High-Dimensional Data (Final Report)

Capital-intensive turbomachinery, such as gas turbines and combined cycle plants, are constantly being monitored for performance anomalies, faults, and physical degradation. Although these power-generating assets are equipped with hundreds of sensors, existing monitoring tools can only handle moderate-sized data. As a result, only a handful of aggregate metrics are used to monitor machine health. At the same time, developing advanced tools suitable for large datasets have been restricted by the lack of appropriate data. The objective of this proposal was to demonstrate a Big Data analytics framework for fault detection and diagnosis in gas turbine applications. We develop a predictive analytics framework methodology guided by these experimental data, industrial data from our collaborators, and physics-based models with engineering domain knowledge. Our analytics framework consists of four key components: (1) a data curation process that addresses data storage, data quality assessments, and integrity checks, (2) a feature engineering component that utilizes statistical methods and transformation algorithms guided by physics-based models to extract high-fidelity fault features that can be leveraged for fault detection and classifying fault severities, (3) a Machine Learning-based fault detection and diagnostics algorithms for detecting operational and hardware faults in the combustion and the turbines section. We utilize two industry-class gas turbine component test rigs to generate first of its kind data for critical gas turbine faults with varying severity levels. Advanced gas turbine test facilities will be interrogated using state-of-the-art instrumentation techniques to build fault signatures and data trends for key combustor and turbine faults. Data generated from a combustor test rig (Georgia Tech) and a turbine test rig (Penn State) during both normal operation and with seeded faults serve as the basis for the Big Data sets. The test conditions in the two test facilities include common, critical events that occur in the operation. Utilizing the combustor test rig, we examine two common combustor faults: lean blowout and centerbody degradation. For the turbine section we develop analytic models for monitoring cooling faults in the gas turbine.

20 FOSSIL-FUELED POWER PLANTS↗
DOE OSTI2021

Real-Time Health Monitoring for Gas Turbine Components Using Online Learning and High-Dimensional Data

Capital-intensive turbomachinery, such as gas turbines and combined cycle plants, are constantly being monitored for performance anomalies, faults, and physical degradation. Although these power-generating assets are equipped with hundreds of sensors, existing monitoring tools can only handle moderate-sized data. As a result, only a handful of aggregate metrics are used to monitor machine health. At the same time, developing advanced tools suitable for large datasets have been restricted by the lack of appropriate data. The objective of this proposal was to demonstrate a Big Data analytics framework for fault detection and diagnosis in gas turbine applications. We develop a predictive analytics framework methodology guided by these experimental data, industrial data from our collaborators, and physics-based models with engineering domain knowledge. Our analytics framework consists of four key components (1) a data curation process that addresses data storage, data quality assessments, and integrity checks, (2) a feature engineering component that utilizes statistical methods and transformation algorithms guided by physics-based models to extract high-fidelity fault features that can be leveraged for fault detection and classifying fault severities, (3) a Machine Learning-based fault detection and diagnostics algorithms for detecting operational and hardware faults in the combustion and the turbines section. We utilize two industry-class gas turbine component test rigs to generate first of its kind data for critical gas turbine faults with varying severity levels. Advanced gas turbine test facilities will be interrogated using state-of-the-art instrumentation techniques to build fault signatures and data trends for key combustor and turbine faults. Data generated from a combustor test rig (Georgia Tech) and a turbine test rig (Penn State) during both normal operation and with seeded faults serve as the basis for the Big Data sets. The test conditions in the two test facilities include common, critical events that occur in the operation. Utilizing the combustor test rig, we examine two common combustor faults: lean blowout and centerbody degradation. For the turbine section we develop analytic models for monitoring cooling faults in the gas turbine

03 NATURAL GAS↗
DOE OSTI2025

Embedded Sensing in Additive Manufacturing Metal and Polymer Parts: A Comparative Study of Integration Techniques and Structural Health Monitoring Performance

This study presents a comparative evaluation of post-process sensor integration in additively manufactured (AM) metal and the in-situ process for polymer structures for structural health monitoring (SHM), with an emphasis on embedded sensors. Geometrically identical specimens were fabricated using copper via metal fused filament fabrication (FFF) and PLA via polymer FFF, with piezoelectric transducers (PZTs) inserted into internal cavities to assess the influence of material and placement on sensing fidelity. Mechanical testing under compressive and point loads generated signals that were transformed into time–frequency spectrograms using a Short-Time Fourier Transform (STFT) framework. An engineered RGB representation was developed, combining global amplitude scaling with an amplitude-envelope encoding to enhance contrast and highlight subtle wave features. These spectrograms served as inputs to convolutional neural networks (CNNs) for classification of load conditions and detection of damage-related features. Results showed reliable recognition in both copper and PLA specimens, with CNN classification accuracies exceeding 95%. Embedded PZTs were especially effective in PLA, where signal damping and environmental sensitivity often hinder surface-mounted sensors. This work demonstrates the advantages of embedded sensing in AM structures, particularly when paired with spectrogram-based feature engineering and CNN modeling, advancing real-time SHM for aerospace, energy, and defense applications.

additive manufacturing↗
NASA NTRS1994

High-Temperature Microphone With Fiber-Optic Output

Acoustic-pressure transducer (microphone) with fiber-optic output designed to withstand hot, loud, structurally vibrating environment like that of jet engine. Features flat frequency response out to frequencies well beyond several-kilohertz range needed to test for acoustic-pressure loads on engine structures.

Hellbaum, Richard F.↗
DOE OSTI2024

ATAT: Astronomical Transformer for time series and Tabular data

Context. The advent of next-generation survey instruments, such as theVera C. RubinObservatory and its Legacy Survey of Space and Time (LSST), is opening a window for new research in time-domain astronomy. The Extended LSST Astronomical Time-Series Classification Challenge (ELAsTiCC) was created to test the capacity of brokers to deal with a simulated LSST stream. Aims. Our aim is to develop a next-generation model for the classification of variable astronomical objects. We describe ATAT, the Astronomical Transformer for time series And Tabular data, a classification model conceived by the ALeRCE alert broker to classify light curves from next-generation alert streams. ATAT was tested in production during the first round of the ELAsTiCC campaigns. Methods. ATAT consists of two transformer models that encode light curves and features using novel time modulation and quantile feature tokenizer mechanisms, respectively. ATAT was trained on different combinations of light curves, metadata, and features calculated over the light curves. We compare ATAT against the current ALeRCE classifier, a balanced hierarchical random forest (BHRF) trained on human-engineered features derived from light curves and metadata. Results. When trained on light curves and metadata, ATAT achieves a macro F1 score of 82.9 ± 0.4 in 20 classes, outperforming the BHRF model trained on 429 features, which achieves a macro F1 score of 79.4 ± 0.1. Conclusions. The use of transformer multimodal architectures, combining light curves and tabular data, opens new possibilities for classifying alerts from a new generation of large etendue telescopes, such as theVera C. RubinObservatory, in real-world brokering scenarios.

Astronomy & Astrophysics↗
NASA NTRS2012

Hydrocarbon Rocket Technology Impact Forecasting

Ever since the Apollo program ended, the development of launch propulsion systems in the US has fallen drastically, with only two new booster engine developments, the SSME and the RS-68, occurring in the past few decades.1 In recent years, however, there has been an increased interest in pursuing more effective launch propulsion technologies in the U.S., exemplified by the NASA Office of the Chief Technologist s inclusion of Launch Propulsion Systems as the first technological area in the Space Technology Roadmaps2. One area of particular interest to both government agencies and commercial entities has been the development of hydrocarbon engines; NASA and the Air Force Research Lab3 have expressed interest in the use of hydrocarbon fuels for their respective SLS Booster and Reusable Booster System concepts, and two major commercially-developed launch vehicles SpaceX s Falcon 9 and Orbital Sciences Antares feature engines that use RP-1 kerosene fuel. Compared to engines powered by liquid hydrogen, hydrocarbon-fueled engines have a greater propellant density (usually resulting in a lighter overall engine), produce greater propulsive force, possess easier fuel handling and loading, and for reusable vehicle concepts can provide a shorter turnaround time between launches. These benefits suggest that a hydrocarbon-fueled launch vehicle would allow for a cheap and frequent means of access to space.1 However, the time and money required for the development of a new engine still presents a major challenge. Long and costly design, development, testing and evaluation (DDT&E) programs underscore the importance of identifying critical technologies and prioritizing investment efforts. Trade studies must be performed on engine concepts examining the affordability, operability, and reliability of each concept, and quantifying the impacts of proposed technologies. These studies can be performed through use of the Technology Impact Forecasting (TIF) method. The Technology Impact Forecasting method is a normative forecasting technique that allows the designer to quantify the effects of adding new technologies on a given design. This method can be used to assess and identify the necessary technological improvements needed to close the gap that exists between the current design and one that satisfies all constraints imposed on the design. The TIF methodology allows for more design knowledge to be brought to the earlier phases of the design process, making use of tools such as Quality Function Deployments, Morphological Matrices, Response Surface Methodology, and Monte Carlo Simulations.2 This increased knowledge allows for more informed decisions to be made earlier in the design process, resulting in shortened design cycle time. This paper will investigate applying the TIF method, which has been widely used in aircraft applications, to the conceptual design of a hydrocarbon rocket engine. In order to reinstate a manned presence in space, the U.S. must develop an affordable and sustainable launch capability. Hydrocarbon-fueled rockets have drawn interest from numerous major government and commercial entities because they offer a low-cost heavy-lift option that would allow for frequent launches1. However, the development of effective new hydrocarbon rockets would likely require new technologies in order to overcome certain design constraints. The use of advanced design methods, such as the TIF method, enables the designer to identify key areas in need of improvement, allowing one to dial in a proposed technology and assess its impact on the system. Through analyses such as this one, a conceptual design for a hydrocarbon-fueled vehicle that meets all imposed requirements can be achieved.

Stuber, Eric↗
NASA NTRS1988

Life prediction modeling based on strainrange partitioning

Strainrange partitioning (SRP) is an integrated low-cycle-fatigue life predicting system. It was created specifically for calculating cyclic crack initiation life under severe high-temperature fatigue conditions. The key feature of the SRP system is its recognition of the interacting mechanisms of cyclic inelastic deformation that govern cyclic life at high temperatures. The SRP system bridges the gap between the mechanistic level of understanding that breeds new and better materials and the phenomenological level wherein workable engineering life prediction methods are in great demand. The system was recently expanded to address engineering fatigue problems in the low-strain, long-life, nominally elastic regime. This breakthrough, along with other advances in material behavior and testing technology, has permitted the system to also encompass low-strain thermomechanical loading conditions. Other important refinements of the originally proposed method include procedures for dealing with life-reducing effects of multiaxial loading, ratcheting, mean stresses, nonrepetitive (cumulative loading) loading, and environmental and long-time exposure. Procedure were also developed for partitioning creep and plastic strain and for estimating strainrange versus life relations from tensile and creep rupture properties. Each of the important engineering features of the SRP system are discussed and examples shown of how they help toward predicting high-temperature fatigue life under practical, although complex, loading conditions.

Halford, Gary R.↗
NASA NTRS1981

NASA VCE test bed engine aerodynamic performance characteristics and test results

The Core Driven Fan Stage (CDFS) Variable Cycle Engine (VCE) has been identified as a leading candidate for advanced supersonic cruise aircraft. A scale demonstrator version of this engine has been designed and tested. This testbed engine features a split fan with double bypass capability, variable forward and aft mixers, and a variable area low pressure turbine nozzle to permit exploration and optimization of the cycle in both single and double bypass modes. This paper presents the aerodynamic performance characteristics and experimental results obtained from both the core engine and full engine tests.

French, M. W.↗
DOE OSTI2025

Enhancing Automotive Intrusion Detection Through Multi-Modal Fusion: A CAN FD-LiDAR Approach

As vehicles become smarter and more autonomous, they increasingly depend on advanced sensors and communication technologies to operate securely. However, such growing dependence on technology—whether it’s CAN (Controller Area Network) for internal communication or LiDAR (Light Detection and Ranging) for sensing the world around them—also expands the attack surface for the types of cyber attacks. Traditional intrusion detection systems (IDS) typically monitor these systems in isolation, limiting their ability to detect sophisticated, crosssystem attacks. To address this, we propose a multi-modal fusion approach that combines real-world CAN FD signals (from the HCRL dataset) with LiDAR features (from the nuScenes dataset) to enhance attack detection. Our method employs a twostage ensemble approach. Calibrated XGBoost and LightGBM models initially process CAN FD (Fuzzing Data) and LiDAR data independently, detecting timing anomalies and space abnormalities. They are subsequently logarithmically combined with a logistic regression meta-model along with 17 engineered features capturing cross-modal behavior, prediction conflicts, and nonlinear interactions. This approach achieves an AUC of 0.87 and an F1-score of 0.82, surpassing single-modality baselines and early fusion methods, at merely 2 ms inference latency. Compared with deep learning competitors, it is 3 times more efficient, providing a lightweight, interpretable, and real time solution to automotive cybersecurity.

97 MATHEMATICS AND COMPUTING↗