Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “temporal learning”

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 523 records · Page 29

State Identification for Planetary Rovers: Learning and Recognition

A planetary rover must be able to identify states where it should stop or change its plan. With limited and infrequent communication from ground, the rover must recognize states accurately. However, the sensor data is inherently noisy, so identifying the temporal patterns of data that correspond to interesting or important states becomes a complex problem. In this paper, we present an approach to state identification using second-order Hidden Markov Models. Models are trained automatically on a set of labeled training data; the rover uses those models to identify its state from the observed data. The approach is demonstrated on data from a planetary rover platform.

Aycard, Olivier↗

Deep Cellular Recurrent Network for Efficient Analysis of Time-Series Data With Spatial Information

Efficient processing of large-scale time series data is an intricate problem in machine learning. Conventional sensor signal processing pipelines with hand engineered feature extraction often involve huge computational cost with high dimensional data. Deep recurrent neural networks have shown promise in automated feature learning for improved time-series processing. However, generic deep recurrent models grow in scale and depth with increased complexity of the data. This is particularly challenging in presence of high dimensional data with temporal and spatial characteristics. Consequently, this work proposes a novel deep cellular recurrent neural network (DCRNN) architecture to efficiently process complex multi-dimensional time series data with spatial information. Here, the cellular recurrent architecture in the proposed model allows for location-aware synchronous processing of time series data from spatially distributed sensor signal sources. Extensive trainable parameter sharing due to cellularity in the proposed architecture ensures efficiency in the use of recurrent processing units with high-dimensional inputs. This study also investigates the versatility of the proposed DCRNN model for classification of multi-class time series data from different application domains. Consequently, the proposed DCRNN architecture is evaluated using two time-series datasets: a multichannel scalp EEG dataset for seizure detection, and a machine fault detection dataset obtained in-house. The results suggest that the proposed architecture achieves state-of-the-art performance while utilizing substantially less trainable parameters when compared to comparable methods in the literature.

60 APPLIED LIFE SCIENCES↗

High-Throughput Microfluidics Platform for Intracellular Delivery and Sampling of Biomolecules from Live Cells

Nondestructive cell membrane permeabilization systems enable the intracellular delivery of exogenous biomolecules for cell engineering tasks as well as the temporal sampling of cytosolic contents from live cells for the analysis of dynamic processes. Here, in this paper, we report a microwell array format live-cell analysis device (LCAD) that can perform localized-electroporation induced membrane permeabilization, for cellular delivery or sampling, and directly interfaces with surface-based biosensors for analyzing the extracted contents. We demonstrate the capabilities of the LCAD via an automated high-throughput workflow for multimodal analysis of live-cell dynamics, consisting of quantitative measurements of enzyme activity using self-assembled monolayers for MALDI mass spectrometry (SAMDI) and deep-learning enhanced imaging and analysis. By combining a fabrication protocol that enables robust assembly and operation of multilayer devices with embedded gold electrodes and an automated imaging workflow, we successfully deliver functional molecules (plasmid and siRNA) into live cells at multiple time-points and track their effect on gene expression and cell morphology temporally. Furthermore, we report sampling performance enhancements, achieving saturation levels of protein tyrosine phosphatase activity measured from as few as 60 cells, and demonstrate control over the amount of sampled contents by optimization of electroporation parameters using a lumped model. Lastly, we investigate the implications of cell morphology on electroporation-induced sampling of fluorescent molecules using a deep-learning enhanced image analysis workflow.

59 BASIC BIOLOGICAL SCIENCES↗

Machine learning to discover mineral trapping signatures due to CO 2 injection

Mineral trapping is pursued as a geological CO 2 sequestration (GCS) mechanism because it permanently stores CO 2 in solid phases or minerals. However, CO 2 mineral-trapping mechanisms are poorly understood due to (1) lack of sufficient field and laboratory data characterizing these complex processes, and (2) challenges to develop site-specific reactive-transport models coupling fluid flow and geochemical reactions occurring at various temporal (from milliseconds to years) and spatial (from pore (millimeters) to field (kilometers)) scales. Reactive transport with additional complexities such as heterogeneity can make the simulation outputs even more difficult to interpret because of complex nonlinearity and multi-scale interdependencies. Furthermore, the values of model outputs such as concentrations can vary by several orders of magnitude, making it harder to correlate and characterize the impact of the variables via traditional data interpretation techniques such as exploratory data analyses. Recently, machine learning (ML) has shown promise in feature discovery and in highlighting hidden mechanisms that cannot be obtained by existing data-analytics and statistical methods. In this study, we applied an unsupervised ML approach, non-negative matrix factorization with custom -means clustering (NMF) to the data generated by reactive-transport simulations of GCS. The reactive-transport data consisted of 19 attributes, including four physio-chemical variables (pH, porosity, aqueous CO 2 , and sequestered CO 2 ), six chemical species (K + , Na + , HCO, Ca 2+ , Mg 2+ , Fe 2+ ), and four carbonate minerals (calcite, dolomite, siderite, and ankerite), a feldspar mineral (albite), and four clay minerals (illite, clinochlore, kaolinite, and smectite) over a period of 200 years of simulation time. Furthermore, the simulation data used was for Morrow B sandstone at the Farnsworth hydrocarbon unit in Texas. Data are sampled at two locations within the model domain: (1) at the injection well and (2) 200 m west of the injection well. The injection was performed for a period of 10 years. Using NMF, we estimated the temporal interdependencies among the 19 attributes over a span of 200 years. We found that NMF was able to identify four reaction stages and their dominant attributes; these cannot be directly discerned through traditional visualization (e.g., line plots, Pareto analysis, Glyph-based visualization methods) or exploratory data analysis tools of the simulation data. The four stages were: reactions in the injection phase followed by short-, mid-, and long-term reactions. The NMF analysis also revealed that 10 among the 19 attributes are dominant. These dominant attributes for mineral trapping include calcite, dolomite at injection well, siderite at 200 m away from the injection well, clinochlore, kaolinite, Na + , K + , Ca 2+ , Mg 2+ , pH, and aqeuous CO 2 . Finally, at late times (65–200 years), our results showed that calcite plays a major role in mineral trapping with insignificant contribution from siderite, ankerite, and clay minerals. These findings make the proposed unsupervised ML-model attractive for reactive-transport sensing towards real-time GCS monitoring.

54 ENVIRONMENTAL SCIENCES↗

A Review of Physics-Informed Machine Learning in Fluid Mechanics

Physics-informed machine-learning (PIML) enables the integration of domain knowledge with machine learning (ML) algorithms, which results in higher data efficiency and more stable predictions. This provides opportunities for augmenting—and even replacing—high-fidelity numerical simulations of complex turbulent flows, which are often expensive due to the requirement of high temporal and spatial resolution. In this review, we (i) provide an introduction and historical perspective of ML methods, in particular neural networks (NN), (ii) examine existing PIML applications to fluid mechanics problems, especially in complex high Reynolds number flows, (iii) demonstrate the utility of PIML techniques through a case study, and (iv) discuss the challenges and opportunities of developing PIML for fluid mechanics.

42 ENGINEERING↗

An interpretable machine learning model for advancing terrestrial ecosystem predictions

We apply an interpretable Long Short-Term Memory (iLSTM) network for land-atmosphere carbon flux predictions based on time series observations of seven environmental variables. iLSTM enables interpretability of variable importance and variable-wise temporal importance to the prediction of targets by exploring internal network structures. The application results indicate that iLSTM not only improves prediction performance by capturing different dynamics of individual variables, but also reasonably interprets the different contribution of each variable to the target and its different temporal relevance to the target. This variable and temporal importance interpretation of iLSTM advances terrestrial ecosystem model development as well as our predictive understanding of the system.

Lu, Dan↗

Machine Learning-Based Investigation of the Associations between Residential Power Consumption and Weather Conditions

Machine learning feature selection analyses were conducted to understand the impact of various weather attributes on power electricity demand and its components in the northwest United States. Load profiles data were obtained from the Northwest Energy Efficiency Alliance (NEEA) and processed to yield hourly load profiles with the same temporal resolution as the weather conditions data obtained from the National Oceanic and Atmospheric Administration (NOAA) for representative weather stations. The data were divided into five seasons; for each season, for each load component, the influences of weather factors were evaluated and quantified using cross-correlation, principal component analysis, and mutual information evaluation; then predictive models were developed based on the ranked/screened factors using the regression tree (RT) approach. After multi-fold cross-validation, the optimal complexity/depth of the regress tree models is found to vary for different load components, and the prediction accuracy using RT and weather data can be higher than 80% for heating and refrigeration, or lower than 30% for vent and cooking. The study provides guidance on improvement of load profile generation, or even serves as a standalone predictive tool for approximating load profiles in balance authorities or climate zones where load profile data is absent.

load composite, load profiling, NEEA, mutual infor↗

Inferring protein expression changes from mRNA in Alzheimer’s dementia using deep neural networks

Identifying the molecular systems and proteins that modify the progression of Alzheimer’s disease and related dementias (ADRD) is central to drug target selection. However, discordance between mRNA and protein abundance, and the scarcity of proteomic data, has limited our ability to advance candidate targets that are mainly based on gene expression. Therefore, by using a deep neural network that predicts protein abundance from mRNA expression, here we attempt to track the early protein drivers of ADRD. Specifically, by applying the clei2block deep learning model to 1192 brain RNA-seq samples, we identify protein modules and disease-associated expression changes that were not directly observed at the mRNA level. Moreover, pseudo-temporal trajectory inference based on the predicted proteome became more closely correlated with cognitive decline and hippocampal atrophy compared to RNA-based trajectories. This suggests that the predicted changes in protein expression could provide a better molecular representation of ADRD progression. Furthermore, overlaying clinical traits on protein pseudotime trajectory identifies protein modules altered before cognitive impairment. These results demonstrate how our method can be used to identify potential early protein drivers and possible drug targets for treating and/or preventing ADRD.

60 APPLIED LIFE SCIENCES↗

Adaptive Graph Learning with Transformer for Multi-Reservoir Inflow Prediction

Reservoir inflow prediction is crucial for water resource management, yet existing approaches mainly focus on single-reservoir models that ignore spatial dependencies among interconnected reservoirs. We introduce AdaTrip as an adaptive, time-varying graph learning framework for multi-reservoir inflow forecasting. AdaTrip constructs dynamic graphs where reservoirs are nodes with directed edges reflecting hydrological connections, employing attention mechanisms to automatically identify crucial spatial and temporal dependencies. Evaluation on thirty reservoirs in the Upper Colorado River Basin demonstrates superiority over existing baselines, with improved performance for reservoirs with limited records through parameter sharing. Additionally, AdaTrip provides interpretable attention maps at edge and time-step levels, offering insights into hydrological controls to support operational decision-making. Our code is available at https://github.com/humphreyhuu/AdaTrip.

Hu, Pengfei [ORNL] (ORCID:0009000367130950)↗

A Data-driven Phytotechnology Framework for Identification and Remediation of Leached-Metals-Contaminated Soil Near Coal Ash Impoundments

This project developed and evaluated advanced remote sensing and machine learning approaches to identify, monitor, and address environmental contamination associated with coal combustion residual (CCR) impoundments and landfills at coal-fired power plants. In Phase I, historical and multi-temporal Sentinel-2 satellite imagery, groundwater monitoring data, and environmental variables were integrated to detect vegetation stress potentially caused by toxic metal leaching from coal ash disposal sites. Multiple vegetation and biophysical indices were analyzed to determine their effectiveness in identifying abnormal vegetation growth patterns linked to contamination. Results from case studies conducted at coal ash–impacted power plant sites in North Carolina and Virginia demonstrated that satellite-based vegetation monitoring can serve as an effective early indicator of environmental stress associated with metals such as arsenic, cadmium, cobalt, lead, lithium, radium, and thallium.

01 COAL, LIGNITE, AND PEAT↗

Doppler Radar Profiler for Launch Winds at the Kennedy Space Center (Phase 1a)

The NASA Engineering and Safety Center (NESC) received a request from the, NASA Technical Fellow for Flight Mechanics at Langley Research Center (LaRC), to develop a database from multiple Doppler radar wind profiler (DRWP) sources and develop data processing algorithms to construct high temporal resolution DRWP wind profiles for day-of-launch (DOL) vehicle assessment. This document contains the outcome of Phase 1a of the assessment including Findings, Observations, NESC Recommendations, and Lessons Learned.

Murri, Daniel G.↗

Remote Sensing of Lineage Functional Types for Modeling and Monitoring Biodiversity

Hyperspectral remote sensing has the potential to continuously scale plant function and plant diversity information from landscape to global extents. Numerous studies have indicated that VSWIR (400-2500 nm) reflectance properties of vegetation capture evolutionarily conserved biochemical, structural, and other functional attributes of plant species. Spectral properties conserved in plants provide the opportunity to both 1) aggregate species into lineages with improved classification accuracy and 2) link those lineages directly to plant traits. Full realization of this goal will enable parameterization of Land Surface Models (LSMs) with remotely sensed information, e.g., canopy nitrogen, and better representations of biodiversity and functional diversity in biogeographic studies. In this study, we use hyperspectral AVIRIS data from the 2013 HyspIRI campaign over the Southern Sierra Nevada, California flight box to investigate the potential for incorporating evolutionary thinking into landcover classification. We link the airborne hyperspectral data with vegetation plot data from roughly 1372 surveys and a phylogeny representing 1361 species. We aggregate species into lineages ranging from species level groups down to similar number of Plant Functional Types as often used in LSMs. We assessed the ability of Random Forest and Partial Least Squares Discriminant Analysis to discriminate across these different phylogenetic scales and determine the optimal number of lineages to classify. Although there are some temporal and spatial differences in our training data, our best approaches achieved moderate classification accuracy (Kappa > 0.65). Given an optimal number of lineages, we explored approaches to improve classifications including machine learning and unmixing approaches. This work suggests that lineage-based methods may be a promising way to leverage the huge amounts of data that will come from high resolution and high return interval hyperspectral data planned for the Surface Biology and Geology mission with sparsely sampled existing ground-based ecological data.

Hyperspectral↗

Rapid detection of rare events from in situ X-ray diffraction data using machine learning

High-energy X-ray diffraction methods can non-destructively map the 3D microstructure and associated attributes of metallic polycrystalline engineering materials in their bulk form. These methods are often combined with external stimuli such as thermo-mechanical loading to take snapshots of the evolving microstructure and attributes over time. However, the extreme data volumes and the high costs of traditional data acquisition and reduction approaches pose a barrier to quickly extracting actionable insights and improving the temporal resolution of these snapshots. This article presents a fully automated technique capable of rapidly detecting the onset of plasticity in high-energy X-ray microscopy data. The technique is computationally faster by at least 50 times than the traditional approaches and works for data sets that are up to nine times sparser than a full data set. This new technique leverages self-supervised image representation learning and clustering to transform massive data sets into compact, semantic-rich representations of visually salient characteristics ( e.g. peak shapes). These characteristics can rapidly indicate anomalous events, such as changes in diffraction peak shapes. It is anticipated that this technique will provide just-in-time actionable information to drive smarter experiments that effectively deploy multi-modal X-ray diffraction methods spanning many decades of length scales.

Zheng, Weijian↗

Acoustic Energy Release During the Laboratory Seismic Cycle: Insights on Laboratory Earthquake Precursors and Prediction

Abstract Machine learning can predict the timing and magnitude of laboratory earthquakes using statistics of acoustic emissions. The evolution of acoustic energy is critical for lab earthquake prediction; however, the connections between acoustic energy and fault zone processes leading to failure are poorly understood. Here, we document in detail the temporal evolution of acoustic energy during the laboratory seismic cycle. We report on friction experiments for a range of shearing velocities, normal stresses, and granular particle sizes. Acoustic emission data are recorded continuously throughout shear using broadband piezo‐ceramic sensors. The coseismic acoustic energy release scales directly with stress drop and is consistent with concepts of frictional contact mechanics and time‐dependent fault healing. Experiments conducted with larger grains (10.5 μm) show that the temporal evolution of acoustic energy scales directly with fault slip rate. In particular, the acoustic energy is low when the fault is locked and increases to a maximum during coseismic failure. Data from traditional slide‐hold‐slide friction tests confirm that acoustic energy release is closely linked to fault slip rate. Furthermore, variations in the true contact area of fault zone particles play a key role in the generation of acoustic energy. Our data show that acoustic radiation is related primarily to breaking/sliding of frictional contact junctions, which suggests that machine learning‐based laboratory earthquake prediction derives from frictional weakening processes that begin very early in the seismic cycle and well before macroscopic failure.

58 GEOSCIENCES↗

PERSIANN-Unet: A Global Deep Learning Framework for Near-Real-Time Precipitation Estimation Using Infrared Data

Access to high-quality, high-resolution, near-real-time precipitation data is essential for hydrological and meteorological research and disaster mitigation. Traditional tools such as rain gauges and radar networks, though effective, have limitations, including sparse coverage in remote areas and high operational costs. Satellite data, with its global coverage and high spatial and temporal resolutions, mitigates limitations in coverage. Satellite precipitation products like Hydro Estimator (HE), Integrated Multi-satellitE Retrievals for Global Precipitation Measurement (IMERG), and Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks (PERSIANN) utilize both geosynchronous thermal infrared (IR) and passive microwave (PMW) data in their operation. PMW sensors offer detailed atmospheric profiles but suffer from higher latency, whereas IR sensors provide lower latency but only capture cloud-top information. Despite this constraint, IR data remains attractive for low-latency precipitation estimation. Recent advances in deep learning, particularly convolutional neural networks (CNNs), have further improved satellite precipitation retrievals. This study introduces PERSIANN-Unet (PUnet or PERSIANN V3), a quasi-global algorithm covering 60°N–60°S that combines IR data, monthly climatology, and the UNet architecture to produce half-hourly precipitation estimates at 0.04° resolution. The product is evaluated against HE, IMERG, and PDIR-Now for 2022–2023. Results show that PUnet closely matches its training target, IMERG V07 Final, at the global scale, and performance is further evaluated against Stage IV as a reference over CONUS. Training PUnet on IMERG (2016–2021) leverages a high-quality, integrated PMW IR-gauge precipitation product while developing an IR-based framework not reliant on PMW availability. By operating on a single global image, PUnet avoids tile partitioning and blending steps, reducing edge discontinuities, and produces more spatially consistent precipitation fields across hemispheres.

Phu Nguyen↗

High-Resolution South American Wind Resource Data Downscaled with Generative Machine Learning Conditioned on Near-Surface Observations

High-resolution historical wind data was developed for the entirety of South America using the innovative Super-Resolution for Renewable Resource Data (sup3r) machine learning framework. The publicly available Sup3rWind South America dataset represents a significant advancement in wind resource data generation, leveraging generative machine learning conditioned on near-surface observations from the Meteorological Assimilation Data Ingest System (MADIS) to efficiently and accurately downscale coarse reanalysis data from the European Centre for Medium-Range Weather Forecasts (ERA5). This approach produces fine-scale, spatially and temporally coherent wind and meteorological fields hundreds of times more computationally efficient than traditional numerical weather modeling methods, enabling access to high-fidelity wind information across both continental and offshore regions. Sup3rWind South America builds on the earlier Sup3rWind Ukraine dataset through improvements in model architecture and outputs conditioned on near-surface observation inputs. As with the Ukraine data release, this dataset includes wind speed, wind direction, temperature, relative humidity, and pressure at a horizontal resolution of ~2 km, representing a 15x spatial enhancement relative to the 31 km ERA5 grid. Wind speed and direction are provided at 5-minute resolution, a 12x temporal refinement compared to the hourly ERA5 data, while temperature, relative humidity, and pressure remain at hourly resolution. The data covers all years from 2005 to 2024. Before downscaling, ERA5 inputs were bias-corrected using long-term monthly means and a limited number of quality-controlled observations to align large-scale statistics with regional conditions. The resulting dataset is the first publicly available high-resolution timeseries wind record that provides full spatial coverage of South America. Model validation demonstrates strong agreement with observations across several statistical metrics, consistent with other state-of-the-art high-resolution wind resource datasets. The potential applications of Sup3rWind South America span renewable energy resource assessment, energy system modeling, and grid resilience analysis. The 20-year record and high spatial and temporal resolution support accurate estimation of long-term energy yield and the economic feasibility of potential wind development sites. Continuous coverage across both continental and offshore regions enables comprehensive site prospecting within exclusive economic zones. The 2 km, 5-minute resolution data provide the spatial and temporal variability required for power system simulation, operational planning, and regional risk assessments.

17 WIND ENERGY↗

Adaptive Protection of Scientific Backbone Networks Using Machine Learning

In this article, we propose a new protection scheme for backbone networks to guarantee high service availability. The presented scheme does not require any reconfiguration immediately after the failure (i.e., it is proactive). At the same time, it does not require any reserved backup network resources either. To achieve these seemingly contradictory goals, we utilize the recent advancements in Machine Learning (ML) to implement a network intelligence that periodically re-allocates the unused capacity as protection bandwidth to meet the service availability requirements of each connection. Our goal is achieved by two components (1) predicting the traffic for the next period on each link, and (2) intelligently selecting the best fit dedicated protection scheme for the next period depending on the estimated unused (spare) bandwidth and the previous service availability violations. Note that re-allocating protection bandwidth affects neither the operational connections nor the current best practice of operators to over-provision network bandwidth to support elephant flows. Finally, we provide a case study on the real traffic from Energy Sciences Network (ESnet), a high-speed, international scientific backbone network. The key benefit of our framework is that adaptively utilizing the over-provisioned bandwidth for spare capacity is sufficient to improve the availability from three-nines to five-nines (in ESnet for the 30 examined connections). The drawback is negligible bandwidth limitations; the user perceives a minor and very temporal bandwidth limitation in less than 0.1% of the time.

42 ENGINEERING↗

Predicting Pulsed-Laser Deposition SrTiO 3 Homoepitaxy Growth Dynamics Using High-Speed Reflection High-Energy Electron Diffraction

Pulsed-laser deposition (PLD) is a powerful technique for growing complex oxides with controlled stoichiometry. To understand growth dynamics therein, it is common to leverage in situ spectroscopies, such as reflection high-energy electron diffraction (RHEED), to monitor surface crystallinity. Most commercial systems rely on video-rate cameras operating at 60-120 Hz that lack sufficient temporal resolution to capture growth dynamics at practical deposition frequencies. Here, a high-speed platform to record in situ dynamics via RHEED at >500 Hz is implemented. An open-source analysis package is designed to fit diffraction spots to 2D Gaussians, allowing single-pulse surface reconstruction kinetics extraction. Using homoepitaxially deposited (001)-oriented SrTiO 3 as a model system, we demonstrate how high-speed RHEED can provide real-time insight into growth processes obscured by slower acquisition systems. By fitting the single-pulse intensity to a set of exponential functions, we observe changes in the characteristic decay time and mechanism correlated to the substrate step width and surface termination. We observe distinct surface effects, with diffraction intensity decaying on lower-energy TiO 2 -terminated surfaces and stabilizing on SrO- or mixed-terminated surfaces. Similarly, using an exponential model, the extracted characteristic time of adatom deposition decreases with increased density of bonding sites associated with mixed termination and narrower step widths. Ultimately, this work shows how increasing RHEED temporal resolution can uncover new insights into growth processes, with practical implications for the design and control of PLD processes. This experimental platform provides new capabilities to enable data-driven machine learning analysis and autonomous control systems to enhance the complexity and fecundity of PLD.

(SrO)↗