Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Dynamic feature selection”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Short-term apartment-level load forecasting using a modified neural network with selected auto-regressive features

Residential electricity load profiles and their diversity have become increasingly important to realize the benefits of Smart or Transactive Energy Networks (TENs). An important element of TENs will be practical, accurate, and implementable residential load forecasting techniques. While there have been many approaches to short-term load forecasting, few have included forecasting for individual households, partly because the high volatility and idiosyncrasies present in individual household load data can pose significant challenges. In this study, we develop a Convolutional Long Short-Term Memory-based neural network with Selected Autoregressive Features (termed a CLSAF model) to improve short-term household electricity load forecasting accuracy by employing three strategies: autoregressive features selection, exogenous features selection, and a “default” state to avoid overfitting at times of high load volatility. We include aggregations of apartments to floor and building level, because utilities may favor transactive approaches that rely on aggregator models, e.g., a cluster of consumers as opposed to an individual. We demonstrate that the CLSAF model, by virtue of its enhanced feature representation and modest computational resources, can accomplish load forecasting in a multi-family residential building across three spatial granularities (individual apartment/household, floor, and building levels), with an accuracy improvement of up to 25% compared to a persistence model. We propose a data screening technique to characterize time-series electricity-load data. This technique is suitable for integration into a TEN ecosystem and allows one to estimate confidence levels of the load forecasts to optimize computational resources and the risks associated with uncertain forecasts.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2022

Background-Aware 3-D Point Cloud Segmentation With Dynamic Point Feature Aggregation

With the proliferation of LiDAR sensors and 3-D vision cameras, 3-D point cloud analysis has attracted significant attention in recent years. In this article, we propose a novel 3-D point cloud learning network, referred to as dynamic point feature aggregation network (DPFA-Net), by selectively performing the neighborhood feature aggregation (FA) with dynamic pooling and an attention mechanism. DPFA-Net has two variants for semantic segmentation and classification of 3-D point clouds. As the core module of the DPFA-Net, we propose an FA layer, in which features of the dynamic neighborhood of each point are aggregated via a self-attention mechanism. In contrast to other segmentation models, which aggregate features from fixed neighborhoods, our approach can aggregate features from different neighbors in different layers providing a more selective and broader view to the query points and focusing more on the relevant features in a local neighborhood. In addition, to further improve the performance of semantic segmentation, we exploit the background–foreground (BF) information and present two novel approaches, namely, two-stage BF-Net and BF regularization. Experimental results show that the proposed DPFA-Net achieves the state-of-the-art overall accuracy score of 89.22% for semantic segmentation on the Stanford large-scale 3-D Indoor Spaces (S3DIS) dataset and provides consistently satisfactory performance across different tasks of semantic segmentation, part segmentation, and 3-D object classification. Furthermore, our model achieves 93.1% accuracy on the ModelNet40 dataset and provides a mean shape intersection-over-union (IoU) value of 85.5% for part segmentation on the ShapeNet-Part dataset. It is a also computationally more efficient compared to other methods.

3-D↗
DOE OSTI2023

Design and implementation of I/O performance prediction scheme on HPC systems through large-scale log analysis

Abstract Large-scale high performance computing (HPC) systems typically consist of many thousands of CPUs and storage units used by hundreds to thousands of users simultaneously. Applications from large numbers of users have diverse characteristics, such as varying computation, communication, memory, and I/O intensity. A good understanding of the performance characteristics of each user application is important for job scheduling and resource provisioning. Among these performance characteristics, I/O performance is becoming increasingly important as data sizes rapidly increase and large-scale applications, such as simulation and model training, are widely adopted. However, predicting I/O performance is difficult because I/O systems are shared among all users and involve many layers of software and hardware stack, including the application, network interconnect, operating system, file system, and storage devices. Furthermore, updates to these layers and changes in system management policy can significantly alter the I/O behavior of applications and the entire system. To improve the prediction of the I/O performance on HPC systems, we propose integrating information from several different system logs and developing a regression-based approach to predict the I/O performance. Our proposed scheme can dynamically select the most relevant features from the log entries using various feature selection algorithms and scoring functions, and can automatically select the regression algorithm with the best accuracy for the prediction task. The evaluation results show that our proposed scheme can predict the write performance with up to 90% prediction accuracy and the read performance with up to 99% prediction accuracy using the real logs from the Cori supercomputer system at NERSC.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

ADDITIVELY MANUFACTURED SURFACE HEAT TRANSFER ENHANCEMENTS FOR THE TRANSFORMATIONAL CHALLENGE REACTOR

The Transformational Challenge Reactor (TCR) is a high-temperature gas-cooled reactor design that uses additively manufactured fuel elements. TCR fuel elements have walls made of silicon carbide and are filled with tristructural-isotropic fuel particles. These fuel elements can have radically different shapes and integrated features than existing designs due to the reduced cost for complex structures in binder jet additive manufacturing. As such, binder jet additive manufacturing enables wall surface features to be embedded that can deliver superior heat transfer performance than smooth wall designs. In this work, the authors conducted a computational fluid dynamics study to evaluate selected wall features integrated into TCR fuel elements. Results show that surface features can outperform smooth wall designs; however, there are unique challenges for gas-cooled reactor core designs that have not been fully explored by previous research. For example, the rough surface finish and process variability of ceramics additive manufacturing make it challenging to predict surface roughness effects before fabrication. Additionally, the small hydraulic diameters of coolant channels in reactor cores make it difficult to engineer surface features that do not significantly increase the pressure drop. Engineers must carefully size surface features for heat transfer enhancement in additive fuel elements to operate above the base material's surface roughness effects and below the coolant channel size.

Weinmeister, Justin↗
DOE OSTI2021

A Confidence-Guided Technique for Tracking Time-Varying Features

Application scientists often employ feature tracking algorithms to capture the temporal evolution of various features in their simulation data. However, as the complexity of the scientific features is increasing with the advanced simulation modeling techniques, quantification of reliability of the feature tracking algorithms is becoming important. One of the desired requirements for any robust feature tracking algorithm is to estimate its confidence during each tracking step so that the results obtained can be interpreted without any ambiguity. To address this, we develop a confidence-guided feature tracking algorithm that allows reliable tracking of user-selected features and presents the tracking dynamics using a graph-based visualization along with the spatial visualization of the tracked feature. Here, the efficacy of the proposed method is demonstrated by applying it to two scientific datasets containing different types of time-varying features.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Controlling Exsolution Dynamics in High‐Entropy Oxides for Highly Active and Selective Acetylene Semi‐Hydrogenation

Exsolution-derived catalysts feature robust metal–support interactions that enhance catalytic performance; yet achieving precise control over exsolution dynamics in multicomponent oxides remains challenging. In this study, we demonstrate that exsolution behavior in high-entropy oxides (HEOs) can be rationally tuned through coupled lattice- and valence-engineering to create a highly active and selective catalyst for acetylene semi-hydrogenation. Incorporation of Li + into a rock salt-structured HEO (LiNiMgCuZnCoO x and LiHEO) induces local lattice distortion, generates oxygen vacancies, and partially oxidizes Co sites from Co 2+ to Co 3+ , collectively modulating local charge redistribution. This strategy enables facilitated Cu nanoparticle exsolution and alters the exsolution sequence from Cu 0 > Ni 0 > Co 0 in pristine HEO to Cu 0 > Co 0 > Ni 0 in the LiHEO. The resulting catalyst via controlled exsolution exhibits superior activity and ethylene selectivity, outperforming state-of-the-art transition metal systems. This work establishes entropy-enabled lattice and valence engineering as a facile route to programmable exsolution for enhanced catalysis.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Uncertainty-informed selection of CMIP6 Earth System Model subsets for use in multisectoral and impact models

Earth system models (ESMs) and general circulation models (GCMs) are heavily used to provide inputs to sectoral impact and multisector dynamic models, which include representations of energy, water, land, economics, and their interactions. Therefore, representing the full range of model uncertainty, scenario uncertainty, and interannual variability that ensembles of these models capture is critical to the exploration of the future co-evolution of the integrated human–Earth system. The pre-eminent source of these ensembles has been the Coupled Model Intercomparison Project (CMIP). With more modeling centers participating in each new CMIP phase, the size of the model archive is rapidly increasing, which can be intractable for impact modelers to effectively utilize due to computational constraints and the challenges of analyzing large datasets. In this work, we present a method to select a subset of the latest phase, CMIP6, featuring models for use as inputs to a sectoral impact or multisector dynamics models, while prioritizing preservation of the range of model uncertainty, scenario uncertainty, and interannual variability in the full CMIP6 ensemble results. This method is intended to help impact modelers select climate information from the CMIP archive efficiently for use in downstream models that require global coverage of climate information. This is particularly critical for large-ensemble experiments of multisector dynamic models that may be varying additional features beyond climate inputs in a factorial design, thus putting constraints on the number of climate simulations that can be used. We focus on temperature and precipitation outputs of CMIP6 models, as these are two of the most used variables among impact models, and many other key input variables for impacts are at least correlated with one or both of temperature and precipitation (e.g., relative humidity). Besides preserving the multi-model ensemble variance characteristics, we prioritize selecting CMIP6 models in the subset that preserve the very likely distribution of equilibrium climate sensitivity values as assessed by the latest Intergovernmental Panel on Climate Change (IPCC) report. This approach could be applied to other output variables of climate models and, possibly when combined with emulators, offers a flexible framework for designing more efficient experiments on human-relevant climate impacts. It can also provide greater insight into the properties of existing CMIP6 models.

Snyder, Abigail C.↗
DOE OSTI2022

Dynamic logistic regression and variable selection: Forecasting and contextualizing civil unrest

Civil unrest can range from peaceful protest to violent furor, and researchers are working to monitor, forecast, and assess such events to allocate resources better. Twitter has become a real-time data source for forecasting civil unrest because millions of people use the platform as a social outlet. Additionally, daily word counts are used as model features, and predictive terms contextualize the reasons for the protest. To forecast civil unrest and infer the reasons for the protest, we consider the problem of Bayesian variable selection for the dynamic logistic regression model and propose using penalized credible regions to select parameters of the updated state vector. This method avoids the need for shrinkage priors, is scalable to high-dimensional dynamic data, and allows the importance of variables to vary in time as new information becomes available. A substantial improvement in both precision and F1-score using this approach is demonstrated through simulation. Finally, we apply the proposed model fitting and variable selection methodology to the problem of forecasting civil unrest in Latin America. Our dynamic logistic regression approach shows improved accuracy compared to the static approach currently used in event prediction and feature selection.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

MDLoader: A Hybrid Model-Driven Data Loader for Distributed Graph Neural Network Training

Scalable data management is essential for processing large scientific dataset on HPC platforms for distributed deep learning. In-memory distributed storage is preferred for its speed, enabling rapid, random, and frequent data access required by stochastic optimizers. Processes use one-sided or collective communication to fetch remote data, with optimal performance depending on (i) dataset characteristics, (ii) training scale, and (iii) interconnection network. Empirical analysis shows collective communication excels with larger mini-batch sizes and/or fewer processes, whereas one-sided communication outperforms at larger scales. We propose MDLoader, a hybrid in-memory data loader for distributed graph neural network training. MDLoader features a model-driven performance estimator that dynamically selects between one-sided and collective communication at the beginning of training using Tree of Parzen Estimators (TPE). Evaluations on NERSC Perlmutter and OLCF Summit show MDLoader outperforms single-backend loaders by up to 2.83 × and predicts the suitable communication method with 96.3% (Perlmutter) and 94.3% (Summit) success rate.

Bae, Jonghyun↗
DOE OSTI2023

Multilevel Robustness for 2D Vector Field Feature Tracking, Selection and Comparison

Abstract Critical point tracking is a core topic in scientific visualization for understanding the dynamic behaviour of time‐varying vector field data. The topological notion of robustness has been introduced recently to quantify the structural stability of critical points, that is, the robustness of a critical point is the minimum amount of perturbation to the vector field necessary to cancel it. A theoretical basis has been established previously that relates critical point tracking with the notion of robustness, in particular, critical points could be tracked based on their closeness in stability, measured by robustness, instead of just distance proximity within the domain. However, in practice, the computation of classic robustness may produce artifacts when a critical point is close to the boundary of the domain; thus, we do not have a complete picture of the vector field behaviour within its local neighbourhood. To alleviate these issues, we introduce a multilevel robustness framework for the study of 2D time‐varying vector fields. We compute the robustness of critical points across varying neighbourhoods to capture the multiscale nature of the data and to mitigate the boundary effect suffered by the classic robustness computation. We demonstrate via experiments that such a new notion of robustness can be combined seamlessly with existing feature tracking algorithms to improve the visual interpretability of vector fields in terms of feature tracking, selection and comparison for large‐scale scientific simulations. We observe, for the first time, that the minimum multilevel robustness is highly correlated with physical quantities used by domain scientists in studying a real‐world tropical cyclone dataset. Such an observation helps to increase the physical interpretability of robustness.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Agentic artificial intelligence for multistage physics experiments at a large-scale user facility particle accelerator

We present a language-model-driven agentic artificial intelligence (AI) system to autonomously execute multistage physics experiments on a production synchrotron light source. Implemented at the Advanced Light Source particle accelerator, the system translates natural language user prompts into structured execution plans that combine archive data retrieval, control-system channel resolution, automated script generation, controlled machine interaction, and analysis. In a representative machine physics task, we show that preparation time was reduced by 2 orders of magnitude relative to manual scripting even for a system expert, while operator-standard safety constraints were strictly upheld. Core architectural features, plan-first orchestration, bounded tool access, and dynamic capability selection, enable transparent, auditable execution with fully reproducible artifacts. These results establish a blueprint for the safe integration of agentic AI into accelerator experiments and demanding machine physics studies, as well as routine operations, with direct portability across accelerators worldwide and, more broadly, to other large-scale scientific infrastructures.

Accelerator/storage ring control systems↗
DOE OSTI2023

A Review of Sensing Technologies for New, Low Global Warming Potential (GWP), Flammable Refrigerants

Commercial refrigeration systems currently utilize refrigerants with global warming potential (GWP) values ranging from 1250 to 4000. The advent of low GWP alternatives (GWP <150) is expected to significantly curtail direct emissions from this segment and greatly influence the ongoing electrification and decarbonization efforts. Most of the low GWP alternatives exhibit flammability risk and hence require robust sensing solutions for a reliable and safe operation of the equipment. This review article aims to provide an overview of different sensing mechanisms suitable for potential applications in systems employing flammable refrigerants, particularly those designated as A2L class. A summary of different A2L refrigerants and their properties is provided followed by a broad review of different classes of sensors, their working principle, transduction method, features, advantages, and limitations. Additionally, key performance characteristics of accuracy, selectivity, sensitivity, dynamic characteristic, and durability among other properties are discussed. Finally, areas of improvement and corresponding approaches are suggested for potential sensors in the successful adoption of A2L class refrigerants.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2022

Data-Centric Development of Lignin Structure–Solubility Relationships in Deep Eutectic Solvents Using Molecular Simulations

Lignin is a natural source of aromatic chemicals with significant potential as an abundant, renewable feedstock for value-added products. Deep eutectic solvents (DES)–solvents composed of a hydrogen bond donor (HBD) and acceptor (HBA) in varying ratios–have emerged as a highly tunable class of solvents for lignin solubilization. However, the variety of possible DES compositions and limited molecular-scale understanding of lignin solubility makes solvent selection a challenge without laborious trial-and-error experimentation. To address these challenges, we use classical molecular dynamics (MD) simulations to study the interactions of lignin model compounds with various DES–water systems. Quantitative parameters (descriptors) were calculated by postprocessing the MD results and used to train a regression model that predicts experimentally determined solubilities of lignin model compounds. This approach revealed that the most important descriptors of solubility are the system temperature, solute hydrophilicity, and metrics quantifying hydrogen bonding. Maximizing the interactions between solute–HBD (hydrophobic group), water–HBD (hydrophilic group), and water–HBA molecules led to the highest model compound solubility. Our results support a hydrotropic mechanism in which extensive DES–water hydrogen bonding and favorable HBD interactions with the solute promote high solubility. We applied the regression model derived using model compounds to predict the solubility of representative lignin oligomers. The model predicted lignin oligomers’ solubilities in good agreement with experiments, indicating that the simulations of model compounds can be extended to predict the solubility of larger lignin compounds across a range of solvent compositions and temperatures. Furthermore, these findings provide new molecular-scale insight into lignin solubilization mechanisms and a new method for computationally screening potential solvent systems for lignin valorization.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Coherent band-edge oscillations and dynamic longitudinal-optical phonon mode splitting as evidence for polarons in perovskites

The coherence of collective modes, such as phonons and polarons, and their modulation of electronic states is long sought in complex systems, which is a crosscutting issue in photovoltaics and quantum electronics. In photovoltaic cells and lasers based on metal halide perovskites, the presence of polarons, i.e., photocarriers dressed by the macroscopic motion of charged lattice, assisted by terahertz (THz) longitudinal-optical (LO) phonons, has been intensely studied yet is still debated. This may be key for explaining the remarkable properties of the perovskite materials, e.g., defect tolerance, long charge lifetimes, and diffusion lengths. In this work we use the intense single-cycle THz pulse with peak electric field up to E THz = 1000 kV/cm to drive coherent polaronic band-edge oscillations at room temperature in CH 3 NH 3 PbI 3 (MAPbI 3 ). We reveal the oscillatory behavior is dominated by a specific quantized lattice vibration mode at $ω$ LO ~ 4 THz, which is both dipole and momentum forbidden. THz-driven coherent polaron dynamics exhibits distinguishing features: room temperature coherent oscillations at ω LO longer than 1 ps in both single crystals and thin films, mode-selective modulation of different band-edge states assisted by electron-phonon interaction, and dynamic mode splitting at low temperature due to entropy and anharmonicity of organic cations. Our results demonstrate intense THz-driven coherent band-edge modulation is a powerful probe of electron-lattice coupling phenomena and polaronic quantum control in perovskites.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Evaluation of Leak Detection Technologies for Low Global Warming Potential (GWP), Flammable Refrigerants

Current commercial refrigeration systems use refrigerants with global warming potential (GWP) values ranging from 1250 to 4000. The emergence of low GWP alternatives (GWP <150) is expected to significantly reduce direct emissions in this sector, playing a crucial role in the ongoing electrification and decarbonization initiatives. However, many of these low GWP alternatives pose a flammability risk, necessitating robust sensing solutions to ensure the reliable and safe operation of the equipment. This paper examines various sensing mechanisms suitable for potential applications in systems that employ flammable refrigerants, specifically those designated as A2L class. It provides a summary of A2L refrigerants and their properties, followed by a comprehensive review of sensor classes, covering their working principles, features, advantages, and limitations. Additionally, the article delves into key performance characteristics such as accuracy, selectivity, sensitivity, dynamic characteristics, and durability, among other properties. The article discusses areas for improvement and suggests corresponding approaches for potential sensors in facilitating the successful adoption of flammable refrigerants. Finally, this paper presents the latest findings from experimental evaluation of 5 different sensing principles in detecting the composition variation as a result of various operational conditions. Reliability and sensitivity of the sensor in responding to shifts in true composition and the resultant LFL value is also discussed.

Reshniak, Viktor↗
DOE OSTI2023

Field-scale dynamics of planting dates in the US Corn Belt from 2000 to 2020

Crop planting dates are a dynamic feature of agricultural systems that respond to short- and long-term climate signals, crop and cultivar selection, and technology changes. Planting date records are essential for yield gap analyses, accurate crop modeling, and tracking farmer adaptations to weather and climate change. Although planting dates have high variation at local scales due to heterogeneity in farm resources and decision-making, available long-term data on planting dates is largely restricted to aggregated regional statistics or, at best, satellite-derived datasets with limited spatiotemporal extent and at resolutions unable to distinguish individual fields (> 250 m). Here, we generated retrospective annual field-scale (30 m) planting date maps for both maize and soybeans spanning 2000-2020 across a 12 state region in the United States Corn Belt based on Landsat satellite data and a large ground sample of over 28,000 maize and soybean fields. Using training data from 2015-2020 for model selection, we found that planting date predictions improved with harmonic regression of Landsat data and additional annual weather covariates. The preferred random forests model approximately doubled performance compared to a null model based on state median planting dates, capturing 47% of field-level variation for maize (mean absolute error, MAE = 7.4 days) and 44% for soybeans (MAE = 7.5 days) against held-out ground truth test data for 2008-2014. We also evaluated the full 2000-2020 dataset with state agricultural statistics, finding strong agreement with median planting dates for maize (R 2 = 0.76, MAE = 4.4 days) and slightly lower agreement for soybeans (R 2 = 0.65, MAE = 5.4 days) when aggregated to the state level. We then used this new dataset to analyze environmental determinants of planting dates at a finer-scale than previously possible, controlling for unobserved variation at the sub-state district level. We found that during 2000-2020, each standard deviation increase in rainfall delayed planting by ~ 2.5 days, and fields with higher soil productivity ratings tended to be planted earlier. We did not find meaningful trends over the last two decades in planting dates for maize or soybeans, in contrast to trends towards earlier planting dates late last century and predicted for this period in climate adaptation studies. We hypothesize increases in early season rainfall may have inhibited these shifts towards earlier planting. Remotely sensed planting dates will be a useful tool for yield gap analyses, crop simulation modeling, and ongoing assessment of climate adaptation.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Contrasting Time-Frequency Representations for Unknown Waveform Detection

In real-world applications like spectrum management and interference detection, dealing with unseen electromagnetic waveforms is critical. Although some methods attempt to simulate open set data using generator models, they face challenges in generating synthetic samples for open set while simultaneously selecting an optimal discriminator for accurate classification. This results in difficulties capturing distinctive features across classes, especially in dynamic scenarios where new classes emerge. To detect unseen waveforms, we propose combining time and frequency domain features with cosine similarity loss to enhance feature distinctiveness and enabling more accurate predictions. This approach efficiently captures more comprehensive information than single-domain representations or approaches without cosine loss. Additionally, our model avoids generic feature vectors by extracting class-specific features during training, resulting in improved class representation. The experiment results show that this combined feature approach with cosine loss outperforms single-domain models and improves accuracy by 10\% over models without cosine loss.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2026

Dynamic permeability in metastable droplet interfacial bilayers

Membrane pores are implicated in several critical functions, including cell fusion and the transport of signaling molecules for intercellular communication. However, these structural features are often difficult to probe directly. Droplet interfacial bilayers offer a synthetic platform to study such membrane properties. We develop a theory that links size-selective transport across a metastable membrane with its transient structural properties. The central quantity of our theory is a dynamic permeability that depends on the mechanism of pore growth, which controls the transient distribution of pore sizes in the membrane. We present a mechanical perspective to derive pore growth dynamics and the resulting size distribution for growth via Ostwald ripening and discuss how these dynamics compare to other growth mechanisms such as coalescence and growth through surfactant desorption. We find scaling relations between the transported particle size, the pore growth rate, and the time for a given fraction of particles to cross the membrane, from which one may deduce the dominant mechanism of pore growth, as well as material properties and structural features of the membrane. Finally, we suggest experiments using droplet interfacial bilayers to validate our theoretical predictions.

Sarma, Nivedina A↗