SEARCH · Engineering Papers

Results for “time series prediction”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 271 records · Page 15

All sources NASA NTRS DOE OSTI

DOE OSTI2020

Data-Driven Modeling and Optimization of Building Energy Consumption: a Case Study

Installing sensors and Building Automation Systems (BAS) allows controlling the facility operations while generating data that can be analyzed for model development. This work focuses on data-driven modeling of the building to optimize energy consumption. The City of Orlando aims to reduce its energy consumption so, they provided us access to their BAS for data and studying the operation of its facilities. We selected a mid-size pilot building to conduct data analysis and modeling. We develop an Application Programming Interface (API) to login to the servers and scrape data. The scraped data contains features ranging from environmental conditions to equipment activity. This dataset is a time series so, it's handled in accordance and analyzed to investigate patterns and relations between data points that help choose parameters for predictive models for building and equipment. Finally, the models are optimized to reduce the energy consumption of the facility.

Grover, Divas↗

NASA NTRS2007

Variations in Stratospheric Inorganic Chlorine Between 1991 and 2006

So how quickly will the ozone hole recover? This depends on how quickly the chlorine content (Cl2) of the atmosphere will decline. The ozone hole forms over the Antarctic each southern spring (September and October). The extremely small ozone amounts in the ozone hole are there because of chemical reactions of ozone with chlorine. This chlorine originates largely from industrially produced chlorofluorocarbon (CFC) compounds. An international agreement, the Montreal Protocol, is drastically reducing the amount of chlorine-containing compounds that we are releasing into the atmosphere. To be able to attribute changes in stratospheric ozone to changes in chlorine we need to know the distribution of atmospheric chlorine. However, due to a lack of continuous observations of all the key chlorine gases, producing a continuous time series of stratospheric chlorine has not been achieved to date. We have for the first time devised a technique to make a 17-year time series for stratospheric chlorine that uses the long time series of HCl observations made from several space borne instruments and a neural network. The neural networks allow us to both inter-calibrate the various HCl instruments and to infer the total amount of atmospheric chlorine from HCl. These new estimates of Cl, provide a much needed critical test for current global models that currently predict significant differences in both Cl(sub y) and ozone recovery. These models exhibit differences in their projection of the recovery time and our chlorine content time series will help separate the good from the bad in these projections.

Lary, D. J.↗

DOE OSTI2021

Long Short-Term Memory Networks for Pattern Recognition of Synthetical Complete Earthquake Catalog

Exploring the spatiotemporal distribution of earthquake activity, especially earthquake migration of fault systems, can greatly to understand the basic mechanics of earthquakes and the assessment of earthquake risk. By establishing a three-dimensional strike-slip fault model, to derive the stress response and fault slip along the fault under regional stress conditions. Our study helps to create a long-term, complete earthquake catalog. We modelled Long-Short Term Memory (LSTM) networks for pattern recognition of the synthetical earthquake catalog. The performance of the models was compared using the mean-square error (MSE). Our results showed clearly the application of LSTM showed a meaningful result of 0.08% in the MSE values. Our best model can predict the time and magnitude of the earthquakes with a magnitude greater than Mw = 6.5 with a similar clustering period. These results showed conclusively that applying LSTM in a spatiotemporal series prediction provides a potential application in the study of earthquake mechanics and forecasting of major earthquake events.

58 GEOSCIENCES↗

DOE OSTI2022

INITIAL STUDIES OF CAVITY FAULT PREDICTION AT JEFFERSON LABORATORY

The Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Laboratory is a CW recirculating linac that utilizes over 400 superconducting radio-frequency (SRF) cavities to accelerate electrons up to 12 GeV through 5-passes. Recent work has shown that, given RF signals from a cavity during a fault as input, machine learning approaches can accurately classify the fault type. In this paper we report on initial results of predicting a fault onset using only data prior to the failure event. A data set was constructed using time-series data immediately before a fault (’unstable’) and 1.5 seconds prior to a fault (’stable’) gathered from over 5,000 saved fault events. The data was used to train a binary classifier. The results gave key insights into the behavior of several fault types and provided motivation to investigate whether data prior to a failure event could also predict the type of fault. We discuss our method using a sliding window approach and report on initial results. Recent modifications to the low-level RF control system will provide access to streaming signals and we outline a path forward for leveraging deep learning on streaming data

Vidyaratne, Lasitha↗

DOE OSTI2021

INITIAL STUDIES OF CAVITY FAULT PREDICTION AT JEFFERSON LABORATORY

Vidyaratne, Lasitha↗

DOE OSTI2025

Deep-learning-based domain adaptation for cavity fault prediction at Jefferson Laboratory

Superconducting radio-frequency (SRF) cavities are the core components of the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab, providing high-power electron beams for nuclear physics experiments. The facility comprises 418 SRF cavities, and any fault in these cavities can lead to interruptions in the electron beam supply. Cavity faults are the leading cause of beam trips in CEBAF. Predicting and mitigating those faults before onset can help maintain normal operation. Existing models face challenges in distinguishing between normal and fault signals when changes occur in the underlying time-series data, from changes in control software, operational parameters, or the environment. This work proposes a deep learning domain adaptation model that leverages transfer learning to address fault prediction challenges by improving accuracy. The model is trained and fine-tuned using a dataset collected for faulty and normal operation using a data acquisition system in CEBAF. Our deep learning-based domain adaptation model achieves a prediction accuracy of 89.61% of the fault and normal signals. The developed model effectively predicts normal running signals compared to the baseline approach without domain adaptation. This capacity is essential for the fault prediction task in the CEBAF because of heavily imbalanced data containing vast amounts of normal signals. The model performs well for predicting faults several hundred milliseconds before the fault onset compared to other models where no adaptation is applied. Incorporating deep learning-based domain adaptation techniques will significantly improve the fault prediction performance.

Rahman, Md Monibor [Old Dominion Univ., Norfolk, V↗

DOE OSTI2022

Learning the Temporal Effect in Infrared Thermal Videos With Long Short-Term Memory for Quality Prediction in Resistance Spot Welding

With the advances of sensing technology, in-situ infrared thermal videos can be collected from Resistance Spot Welding (RSW) processes. Each video records the formulation process of a weld nugget. The nugget evolution creates a “temporal effect” across the frames, which can be leveraged for real-time, nondestructive evaluation (NDE) of the weld quality. Currently, quality prediction with imaging data mainly focuses on optical feature extraction with Convolutional Neural Network (CNN) but does not make the most of such temporal effect. In this study, pixels corresponding to critical locations on the weld nugget surface are extracted from a video to form multivariate time series (MTS). Multivariate Adaptive Regression Splines (MARS) is used in MTS processing to remove noisy signals related to uninformative frames. A Stacked Long Short-Term Memory (LSTM) model is developed to learn from the processed MTS and then predicts weld nugget size and thickness in real-time NDE. Results from a case study on RSW of Boron steel demonstrates the improvement in prediction accuracy and computational time with the proposed method, as compared to CNN-based weld quality prediction.

Guo, Shenghan↗

NASA NTRS2023

Maldives Climate II: Evaluating the Potential Impacts of Sea Level Rise on Human Development and Coastal Infrastructure

The Republic of the Maldives is a low-lying island nation in the Indian Ocean which has experienced rapid urbanization, landcover changes, and sea level rise over recent years. The growth of tourism, coastal erosion, and urbanization have all driven land reclamation efforts across many islands. As in-situ landcover change monitoring has proven difficult across the vast archipelago, the NASA DEVELOP team collaborated with the Maldives Ministry of Environment, Climate Change, and Technology; USAID; and the U.S. Department of State, to utilize Earth observations to predict sea level rise impacts on coastal infrastructure. The team used a supervised classification algorithm within Google Earth Engine to create land use maps and time series analyses of nine islands and atolls using imagery from Landsat 7 Enhanced Thematic Mapper Plus, Landsat 8 Operational Land Imager, Sentinel-2 Multispectral Instrument, and PlanetScope, covering a combined period of 2000 through 2023. Additionally, the team projected coastal inundation with a modified deterministic Bathtub model utilizing elevation data from CoastalDEM and 2050-2100 Shared Socioeconomic Pathway scenarios identified in the NASA Sea Level Rise Projection Tool. The team found that islands undergoing urban growth experienced a 23% decrease in vegetation between 2014 and 2022. Furthermore, the model predicted that 51–57% of the study area’s-built environment has a chance of inundation by 2100 under the low and high sea level scenarios. These analyses demonstrate how remote sensing can be used to both track land use changes over time and project how coastlines will be affected by sea level rise.

remote sensing↗

DOE OSTI2021

Statistical learning for predicting density–matrix-based electron dynamics

In this study, we consider the problem of learning density-dependent molecular Hamiltonian matrices from time series of electron density matrices, all in the context of Hartree–Fock theory. Prior work developed a solution to this problem for small molecular systems with density and Hamiltonian matrices of size at most 6 × 6. Here, using a battery of techniques, we scale prior methods to larger molecular systems with, for example, 29 × 29 matrices. This includes systems that either have more electrons or are expressed in large basis sets such as 6-311++G**. Scaling the method to larger systems enhances its relevance for realistic applications in chemistry and physics. To achieve this scaling, we apply dimensionality reduction, ridge regression and analytic computation of Hessians. Through the combination of these techniques, we are able to learn Hamiltonians by minimizing an objective function that encodes local propagation error. Importantly, these learned Hamiltonians can then be used to predict electron dynamics for thousands of steps: When we use our learned Hamiltonians to numerically solve the time-dependent Hartree–Fock equation, we obtain predicted dynamics that are in close quantitative agreement with ground truth dynamics. This includes field-off trajectories similar to the training data and field-on trajectories outside of the training data.

59 BASIC BIOLOGICAL SCIENCES↗

NASA NTRS1979

Vertical motions in the equatorial middle atmosphere

A single station vertical velocity equation which considers ageostrophic and diabatic effects derived from the first law of thermodynamics and a generalized thermal wind relation is presented. An analysis and verification procedure which accounts for measurement and calculation errors as well as time and space continuity arguments and theoretical predictions are described. Vertical velocities are calculated at every kilometer between 25 and 60 km and for approximately every three hours for the above diurnal period at Kourou (French Guiana), Fort Sherman (Panama Canal Zone), Ascension Island, Antigua (British West Indies) and Natal (Brazil). The results, plotted as time series cross sections, suggest vertical motions ranging in magnitude from 1 or 2 cm/sec at 30 km to as much as 15 cm/sec at 60 km. Many of the general features of the results agree well with atmospheric tidal predictions but many particular features suggest that both smaller time scale gravity waves (periods less than 6 hours) and synoptic type waves (periods greater than 1 day) may be interacting significantly with the tidal fields. The results suggest that vertical motions can be calculated for the equatorial middle atmosphere and must be considered a significant part of the motion for time scales from 8 to 24 hours.

Weisman, M. L.↗

NASA NTRS1994

Modeling time-dependent corrosion fatigue crack propagation in 7000 series aluminum alloys

Stress corrosion cracking and corrosion fatigue experiments were conducted with the susceptible S-L orientation of AA7075-T651, immersed in acidified and inhibited NaCl solution, to provide a basis for incorporating environmental effects into fatigue crack propagation life prediction codes such as NASA FLAGRO. This environment enhances da/dN by five to ten-fold compared to fatigue in moist air. Time-based crack growth rates from quasi-static load experiments are an order of magnitude too small for accurate linear superposition prediction of da/dN for loading frequencies above 0.001 Hz. Alternate methods of establishing da/dt, based on rising-load or ripple-load-enhanced crack tip strain rate, do not increase da/dt and do not improve linear superposition. Corrosion fatigue is characterized by two regimes of frequency dependence; da/dN is proportional to f(exp -1) below 0.001 Hz and to F(exp 0) to F(exp -0.1) for higher frequencies. Da/dN increases mildly both with increasing hold-time at K(sub max) and with increasing rise-time for a range of loading waveforms. The mild time-dependence is due to cycle-time-dependent corrosion fatigue growth. This behavior is identical for S-L nd L-T crack orientations. The frequency response of environmental fatigue in several 7000 series alloys is variable and depends on undefined compositional or microstructural variables. Speculative explanations are based on the effect of Mg on occluded crack chemistry and embritting hydrogen uptake, or on variable hydrogen diffusion in the crack tip process zone. Cracking in the 7075/NaCl system is adequately described for life prediction by linear superposition for prolonged load-cycle periods, and by a time-dependent upper bound relationship between da/dN and delta K for moderate loading times.

Mason, Mark E.↗

DOE OSTI2021

EPIsembleVis: A geo-visual analysis and comparison of the prediction ensembles of multiple COVID-19 models

In this work, we present EPIsembleVis, a web-based comparative visual analysis tool for evaluating the consistency of multiple COVID-19 prediction models. Our approach analyzes a collection of COVID-19 predictions from different epidemiological models as an ensemble and utilizes two metrics to quantify model performance. These metrics include (a) prediction uncertainty (represented as the dispersion of predictions in each ensemble) and (b) prediction error (calculated by comparing individual model predictions with the recorded data). Through an interactive visual interface, our approach provides a data-driven workflow for (a) selecting and constructing the COVID-19 model prediction ensemble based on the spatiotemporal overlap of available predictions of multiple epidemiological models, (b) quantifying the model performance using both the uncertainty of each model prediction ensemble, and the error of each ensemble member that represents individual model predictions, and (c) visualizing the spatiotemporal variability in the projection performance of individual models using a suite of novel ensemble visualization techniques, such as the data availability map, a spatiotemporal textured-tile calendar, multivariate rose chart, and time-series leaflet glyph. We demonstrate the capability of our ensemble visual interface through a case study that investigates the performance of weekly COVID-19 predictions, which are provided through the COVID-19 Forecast Hub UMass-Amherst Influenza Forecasting Center of Excellence [47] for the United States and United States Territories. The EPIsembleVis tool is implemented using open-source web technologies and adaptive system design, rendering it interoperable with Elasticsearch and Kibana for automatically ingesting COVID-19 predictions from online repositories, and it is generalizable for analyzing worldwide projections from more epidemiological models.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2022

Detecting technological maturity from bibliometric patterns

We report the capability to identify emergent technologies based upon easily accessed open-source indicators, such as publications, is important for decision-makers in industry and government. The scientific contribution of this work is the proposition of a machine learning approach to the detection of the maturity of emerging technologies based on publication counts. Time-series of publication counts have universal features that distinguish emerging and growing technologies. We train an artificial neural network classifier, a supervised machine learning algorithm, upon these features to predict the maturity (emergent vs. growth) of an arbitrary technology. With a training set comprised of 22 technologies we obtain a classification accuracy ranging from 58.3% to 100% with an average accuracy of 84.6% for six test technologies. To enhance classifier performance, we augmented the training corpus with synthetic time-series technology life cycle curves, formed by calculating weighted averages of curves in the original training set. Training the classifier on the synthetic data set resulted in improved accuracy, ranging from 83.3% to 100% with an average accuracy of 90.4% for the test technologies. The performance of our classifier exceeds that of competing machine learning approaches in the literature, which report an average classification accuracy of only 85.7% at maximum. Moreover, in contrast to current methods our approach does not require subject matter expertise to generate training labels, and it can be automated and scaled.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Emulator-based Bayesian calibration of a subglacial drainage model

Subglacial drainage models, often motivated by the relationship between hydrology and ice flow, sensitively depend on numerous unconstrained parameters. We explore using borehole water-pressure time series to calibrate the uncertain parameters of a popular subglacial drainage model, taking a Bayesian perspective to quantify the uncertainty in parameter estimates and in the calibrated model predictions. To reduce the computation time associated with Markov Chain Monte Carlo sampling, we construct a fast Gaussian process emulator to stand in for the subglacial drainage model. We first carry out a calibration experiment using synthetic observations consisting of model simulations with hidden parameter values as a demonstration of the method. Using real borehole water pressures measured in western Greenland, we find meaningful constraints on four of the eight model parameters and a factor-of-three reduction in uncertainty of the calibrated model predictions. These experiments illustrate Gaussian process-based Bayesian inference as a useful tool for calibration and uncertainty quantification of complex glaciological models using field data. However, significant differences between the calibrated model and the borehole data suggest that structural limitations of the model, rather than poorly constrained parameters or computational cost, remain the most important constraint on subglacial drainage modelling.

58 GEOSCIENCES↗

DOE OSTI2020

Potential impacts of climate change on wind and solar electricity generation in Texas

Wind and solar energy sources are climate and weather dependent, therefore susceptible to a changing climate. in this paper, we quantify the impacts of climate change on wind and solar electricity generation under high concentrations of greenhouse gases in Texas. We employ mid-twenty-first century climate projections and a high-resolution numerical weather prediction model to generate weather variables in the future and produce wind and solar generation time series. We find that mid-twenty-first century projections based on five global climate models agree on the multiyear average increases across Texas in direct normal irradiance, global horizontal irradiance, surface air temperature, and 100-m wind speed of up to 5%, 4%, 10%, and 1%, respectively. These changes lead to multiyear average relative changes across Texas of -0.6 to + 2.5% and of + 1.3 to + 3.5% in solar and wind capacity factors, respectively, with significant regional, seasonal, and diurnal differences. Areas with low solar resource show an increase in solar capacity factors but reductions in wind capacity factors. Areas with high solar resource show reductions in solar capacity factors. The spatial and temporal differences in our results highlight the importance of using high-resolution data sets to study the potential impacts of climate change on wind and solar power.

14 SOLAR ENERGY↗

DOE OSTI2021

Climate Extremes and Variability Surrounding Chesapeake Bay: Past, Present, and Future

Measures of extreme climate variability around Chesapeake Bay have changed over the past century (1895–2014), resulting in the need to establish new baselines for understanding future change. Here, observed climate variability is compared with Coupled Model Intercomparison Project fifth assessment climate models to evaluate ensemble model skill in this region. Observed trends include annual and seasonal declines in the percentage of cold days. Similarly, increases in the annual and seasonal percentages of warm days occurred in all seasons of the North Chesapeake (>38.2°N) but were only significant in spring and summer in the South (<38.2°N) demonstrating regional differences even at local scales. Precipitation intensity increased over the past century. Models using emissions scenarios RCP4.5 and 8.5 project these trends will continue, though they have little skill in precipitation extremes. Pacific and Atlantic climate modes are generally more correlated with climate extremes than mean temperature and precipitation suggesting potential predictability in the synoptic patterns underlying the extremes. However, they also drive the need for time series at least 60-years long for establishing climate trends. This local analysis differs from regional, state level, and local 2.5° × 3.75° grid analyses, highlighting the importance of local climate assessments that consider topographic and regional weather patterns.

54 ENVIRONMENTAL SCIENCES↗

NASA NTRS2007

Dynamic and Regression Modeling of Ocean Variability in the Tide-Gauge Record at Seasonal and Longer Periods

Comparison of monthly mean tide-gauge time series to corresponding model time series based on a static inverted barometer (IB) for pressure-driven fluctuations and a ocean general circulation model (OM) reveals that the combined model successfully reproduces seasonal and interannual changes in relative sea level at many stations. Removal of the OM and IB from the tide-gauge record produces residual time series with a mean global variance reduction of 53%. The OM is mis-scaled for certain regions, and 68% of the residual time series contain a significant seasonal variability after removal of the OM and IB from the tide-gauge data. Including OM admittance parameters and seasonal coefficients in a regression model for each station, with IB also removed, produces residual time series with mean global variance reduction of 71%. Examination of the regional improvement in variance caused by scaling the OM, including seasonal terms, or both, indicates weakness in the model at predicting sea-level variation for constricted ocean regions. The model is particularly effective at reproducing sea-level variation for stations in North America, Europe, and Japan. The RMS residual for many stations in these areas is 25-35 mm. The production of "cleaner" tide-gauge time series, with oceanographic variability removed, is important for future analysis of nonsecular and regionally differing sea-level variations. Understanding the ocean model's strengths and weaknesses will allow for future improvements of the model.

Hill, Emma M.↗

DOE OSTI2026

A Centralized AI Lakehouse Framework for Brain Tumor MRI Classification and Segmentation, University KPI Forecasting, and Water Potability Prediction

In many university and healthcare projects, models are built for very different data types such as tables, institutional time series, and medical images, but they are deployed as separate applications. In this work, that separation made testing and maintenance difficult because each module had its own pipeline and runtime requirements. This paper presents an integrated AI lakehouse-style implementation that runs three model pipelines inside one containerized backend. For medical imaging, we used MRI datasets from IEEE DataPort: a four-class classification set with 7012 images (5708 train/1304 test) and a segmentation set with 3063 image–mask pairs. The classification model (ResNet50 transfer learning) is evaluated using a proper train–validation–test protocol across multiple splits (80/10/10, 70/10/20, 60/10/30, and 10/30/60), achieving a test accuracy of 99.00% under the standard 80/10/10 split. Additionally, a patient-level evaluation is conducted using an external glioma dataset to provide a more realistic assessment without data leakage. The segmentation model (DeepLabV3-ResNet50) achieved 83.09% validation mIoU and 88.79% Dice score. For university KPI forecasting, we used annual IPEDS and NSF HERD data from 2010 to 2023 for three universities (BSU, EOU, and UAB). To examine the effect of preprocessing on forecasting performance, two case studies are conducted. In the first case, linear interpolation is applied to generate semester-level data. In the second case, the original annual data is used directly without interpolation. Random Forest regression and ARIMA models are evaluated using MAE, RMSE, MAPE, and R 2 . The results showed that interpolation improved apparent forecasting performance due to smoothing, while evaluation on the original annual data provided a more realistic assessment of model behavior. To further validate the framework on a larger dataset, an additional case study is conducted using a student dropout dataset. For water potability, we trained and compared multiple tabular classifiers on a large dataset (1,048,575 samples). A Random Forest model (100 trees, max depth 10) achieved 85.86% test accuracy and high recall for unsafe samples (0.8447). All modules are served via FastAPI and deployed together using Docker, with workflow automation routing requests to the correct endpoint. System-level benchmarking indicates that the backend maintains stable throughput and latency under concurrent requests.

97 MATHEMATICS AND COMPUTING↗