Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “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.

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

At least 109 records · Page 6
All sourcesNASA NTRSDOE OSTI
DOE OSTI2018

At the Nexus of History, Ecology, and Hydrobiogeochemistry: Improved Predictions across Scales through Integration

To improve predictions of ecosystem function in future environments, we need to integrate the ecological and environmental histories experienced by microbial communities with hydrobiogeochemistry across scales. A key issue is whether we can derive generalizable scaling relationships that describe this multiscale integration. There is a strong foundation for addressing these challenges. We have the ability to infer ecological history with null models and reveal impacts of environmental history through laboratory and field experimentation. Recent developments also provide opportunities to inform ecosystem models with targeted omics data. A major next step is coupling knowledge derived from such studies with multiscale modeling frameworks that are predictive under non-steady-state conditions. This is particularly true for systems spanning dynamic interfaces, which are often hot spots of hydrobiogeochemical function. We can advance predictive capabilities through a holistic perspective focused on the nexus of history, ecology, and hydrobiogeochemistry.

Stegen, James C. (ORCID:0000000191357424)↗
DOE OSTI2018

Predicting Near-Term Changes in the Earth System: A Large Ensemble of Initialized Decadal Prediction Simulations Using the Community Earth System Model

The objective of near-term climate prediction is to improve our fore-knowledge, from years to a decade or more in advance, of impactful climate changes that can in general be attributed to a combination of internal and externally forced variability. Predictions initialized using observations of past climate states are tested by comparing their ability to reproduce past climate evolution with that of uninitialized simulations in which the same radiative forcings are applied. A new set of decadal prediction (DP) simulations has recently been completed using the Community Earth System Model (CESM) and is now available to the community. This new large-ensemble (LE) set (CESM-DPLE) is composed of historical simulations that are integrated forward for 10 years following initialization on 1 November of each year between 1954 and 2015. CESM-DPLE represents the “initialized” counterpart to the widely studied CESM Large Ensemble (CESM-LE); both simulation sets have 40-member ensembles, and they use identical model code and radiative forcings. Comparing CESM-DPLE to CESM-LE highlights the impacts of initialization on prediction skill and indicates that robust assessment and interpretation of DP skill may require much larger ensembles than current protocols recommend. CESM-DPLE exhibits significant and potentially useful prediction skill for a wide range of fields, regions, and time scales, and it shows widespread improvement over simpler benchmark forecasts as well as over a previous initialized system that was submitted to phase 5 of the Coupled Model Intercomparison Project (CMIP5). Finally, the new DP system offers new capabilities that will be of interest to a broad community pursuing Earth system prediction research.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Data for FUN-PROSE: A Deep Learning Approach to Predict Condition-Specific Gene Expression in Fungi

mRNA levels of all genes in a genome is a critical piece of information defining the overall state of the cell in a given environmental condition. Being able to reconstruct such condition-specific expression in fungal genomes is particularly important to metabolically engineer these organisms to produce desired chemicals in industrially scalable conditions. Most previous deep learning approaches focused on predicting the average expression levels of a gene based on its promoter sequence, ignoring its variation across different conditions. Here we present FUN-PROSE—a deep learning model trained to predict differential expression of individual genes across various conditions using their promoter sequences and expression levels of all transcription factors. We train and test our model on three fungal species and get the correlation between predicted and observed condition-specific gene expression as high as 0.85. We then interpret our model to extract promoter sequence motifs responsible for variable expression of individual genes. We also carried out input feature importance analysis to connect individual transcription factors to their gene targets. A sizeable fraction of both sequence motifs and TF-gene interactions learned by our model agree with previously known biological information, while the rest corresponds to either novel biological facts or indirect correlations.

Genomics↗
DOE OSTI2022

Energy Prediction Impact of the Space Level Occupancy Schedule for a Primary School

By using the same occupant schedule for all spaces, a building level occupancy schedule in building energy modelling can reduce the cost and time of data collection, especially for large-scale simulations or when detailed occupancy data cannot be obtained. However, by describing the unique occupancy status in each space, a space level schedule can better reflect real-world scenarios. This research investigates the energy prediction impact of a space level occupancy schedule for primary school modelling in 16 ASHRAE climate zones. The results show that when switched from a building level to a space level occupancy schedule, the energy prediction difference is between -1.0% and 0.8%. Generally, the predicted building energy consumption using a space level occupancy schedule is higher than when using a building level occupancy schedule in hot and warm climate zones, but lower in other climate zones.

building energy model↗
DOE OSTI2022

Prescreening-Based Subset Selection for Improving Predictions of Earth System Models With Application to Regional Prediction of Red Tide

We present the ensemble method of prescreening-based subset selection to improve ensemble predictions of Earth system models (ESMs). In the prescreening step, the independent ensemble members are categorized based on their ability to reproduce physically-interpretable features of interest that are regional and problem-specific. The ensemble size is then updated by selecting the subsets that improve the performance of the ensemble prediction using decision relevant metrics. We apply the method to improve the prediction of red tide along the West Florida Shelf in the Gulf of Mexico, which affects coastal water quality and has substantial environmental and socioeconomic impacts on the State of Florida. Red tide is a common name for harmful algal blooms that occur worldwide, which result from large concentrations of aquatic microorganisms, such as dinoflagellate Karenia brevis, a toxic single celled protist. We present ensemble method for improving red tide prediction using the high resolution ESMs of the Coupled Model Intercomparison Project Phase 6 (CMIP6) and reanalysis data. The study results highlight the importance of prescreening-based subset selection with decision relevant metrics in identifying non-representative models, understanding their impact on ensemble prediction, and improving the ensemble prediction. These findings are pertinent to other regional environmental management applications and climate services. Additionally, our analysis follows the FAIR Guiding Principles for scientific data management and stewardship such that data and analysis tools are findable, accessible, interoperable, and reusable. As such, the interactive Colab notebooks developed for data analysis are annotated in the paper. This allows for efficient and transparent testing of the results’ sensitivity to different modeling assumptions. Moreover, this research serves as a starting point to build upon for red tide management, using the publicly available CMIP, Coordinated Regional Downscaling Experiment (CORDEX), and reanalysis data.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2017

A Hubble Space Telescope Survey for Novae in M87. II. Snuffing out the Maximum Magnitude–Rate of Decline Relation for Novae as a Non-standard Candle, and a Prediction of the Existence of Ultrafast Novae

The extensive grid of numerical simulations of nova eruptions from the work of Yaron et al. first predicted that some classical novae might significantly deviate from the Maximum Magnitude–Rate of Decline (MMRD) relation, which purports to characterize novae as standard candles. Kasliwal et al. have announced the observational detection of a new class of faint, fast classical novae in the Andromeda galaxy. These objects deviate strongly from the MMRD relationship, as predicted by Yaron et al. Recently, Shara et al. reported the first detections of faint, fast novae in M87. These previously overlooked objects are as common in the giant elliptical galaxy M87 as they are in the giant spiral M31; they comprise about 40% of all classical nova eruptions and greatly increase the observational scatter in the MMRD relation. We use the extensive grid of the nova simulations of Yaron et al. to identify the underlying causes of the existence of faint, fast novae. These are systems that have accreted, and can thus eject, only very low-mass envelopes, of the order of 10{sup −7}–10{sup −8} M {sub ⊙}, on massive white dwarfs. Such binaries include, but are not limited to, the recurrent novae. These same models predict the existence of ultrafast novae that display decline times, t {sub 2,} to be as short as five hours. We outline a strategy for their future detection.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1988

Lagged average predictions in a predictability experiment

Lagged average predictions are examined here within the context of an idealized predictability experiment. Lagged predictions contribute to making better forecasts than the forecasts obtained from using only the latest initial state. Analytic models suggest that lagged predictions contribute the greatest amount when the error growth rates are small. Little dependence upon the magnitude of the intial error is found if the growth rates remain constant. It is also shown how lagged average forecasts can be used to predict the error. Discriminating forecasts made only when the error is predicted to be small are shown to have much better than average skill.

Roads, John O.↗
DOE OSTI2018

A program for automatically predicting supramolecular aggregates and its application to urea and porphin [A programme for the automated geometry prediction of supra-molecular aggregates and its application to the examples of urea and porphin]

Not only the molecular structure but also the presence or absence of aggregates determines many properties of organic materials. Theoretical investigation of such aggregates requires the prediction of a suitable set of diverse structures. Here, we present the open–source program EnergyScan for the unbiased prediction of geometrically diverse sets of small aggregates. Its bottom–up approach is complementary to existing ones by performing a detailed scan of an aggregate's potential energy surface, from which diverse local energy minima are selected. We crossvalidate this approach by predicting both literature–known and heretofore unreported geometries of the urea dimer. We also predict a diverse set of dimers of the less intensely studied case of porphin, which we investigate further using quantum chemistry. For several dimers, we find strong deviations from a reference absorption spectrum, which we explain using computed transition densities. Furthermore, this proof of principle clearly shows that EnergyScan successfully predicts aggregates exhibiting large structural and spectral diversity.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

On Building Predictive Digital Twin Incorporating Wave Predicting Capabilities: Case Study on UMaine Experimental Campaign - FOCAL

The response of floating wind turbines (FWT) are susceptible to stochastic wave variations. For the optimal operation of FWT, a comprehensive understanding of the phaseresolved wave dynamics and the consequential system response is crucial for real-time monitoring and control. A multi-variate, multi-step, long short term memory (MLSTM), a type of recurrent neural network (RNN) is used to capture complex system dynamics for real-time application. Results indicate that the integration of a wave prediction-reconstruction (WRP) model substantially enhances prediction accuracy by 50% on average relative to the baseline model. The improvement is consistent across various wave extremity and prediction horizons, thereby significantly broadening the scope for timely and precise predictive capabilities.

17 WIND ENERGY↗
DOE OSTI2023

Predicting wind farm operations with machine learning and the P2D‐RANS model: A case study for an AWAKEN site

Abstract The power performance and the wind velocity field of an onshore wind farm are predicted with machine learning models and the pseudo‐2D RANS model, then assessed against SCADA data. The wind farm under investigation is one of the sites involved with the American WAKE experimeNt (AWAKEN). The performed simulations enable predictions of the power capture at the farm and turbine levels while providing insights into the effects on power capture associated with wake interactions that operating upstream turbines induce, as well as the variability caused by atmospheric stability. The machine learning models show improved accuracy compared to the pseudo‐2D RANS model in the predictions of turbine power capture and farm power capture with roughly half the normalized error. The machine learning models also entail lower computational costs upon training. Further, the machine learning models provide predictions of the wind turbulence intensity at the turbine level for different wind and atmospheric conditions with very good accuracy, which is difficult to achieve through RANS modeling. Additionally, farm‐to‐farm interactions are noted, with adverse impacts on power predictions from both models.

17 WIND ENERGY↗
DOE OSTI2017

Evaluating penalized logistic regression models to predict Heat-Related Electric grid stress days

Understanding the conditions associated with stress on the electricity grid is important in the development of contingency plans for maintaining reliability during periods when the grid is stressed. In this paper, heat-related grid stress and the relationship with weather conditions is examined using data from the eastern United States. Penalized logistic regression models were developed and applied to predict stress on the electric grid using weather data. The inclusion of other weather variables, such as precipitation, in addition to temperature improved model performance. Several candidate models and datasets were examined. A penalized logistic regression model fit at the operation-zone level was found to provide predictive value and interpretability. Additionally, the importance of different weather variables observed at different time scales were examined. Maximum temperature and precipitation were identified as important across all zones while the importance of other weather variables was zone specific. The methods presented in this work are extensible to other regions and can be used to aid in planning and development of the electrical grid.

Bramer, L. M.↗
DOE OSTI2017

Predictive modeling of crystal accumulation in high-level waste glass melters processing radioactive waste

The effectiveness of HLW vitrification is limited by precipitation/accumulation of spinel crystals [(Fe, Ni, Mn, Zn)(Fe, Cr)2O4] in the glass discharge riser of Joule-heated ceramic melters during idling. These crystals do not affect glass durability; however, if accumulated in thick layer, they can clog the melter and prevent discharge of molten glass into canisters. To address this problem, an empirical model was developed that can predict thicknesses of accumulated layers as a function of glass composition. This model predicts well the accumulation of single crystals and/or small-scale agglomerates, but, excessive agglomeration observed in high-Ni-Fe glass resulted in an under-prediction of accumulated layers, which gradually worsen over time as an increased number of agglomerates formed. Accumulation rate of ~53.8 ± 3.7 µm/h determined for this glass will result in ~26 mm thick layer in 20 days of melter idling.

Crystal accumulation↗
DOE OSTI2021

Predictive Synthesis

Current solid state synthesis intrinsically involves a multidimensional space which is challenging to parametrize and predict. The diversity of extended structures comes from the diversity of basic properties of elements of the Periodic system which may exhibit a variety of bonding modes. The fundamental challenges of prediction of the preparative outcome are further complicated by practical synthetic issues. Current development in computational and experimental methods calls for collaborative efforts to make solid state synthesis more predictable. Overall, this Perspective discusses several steps in this direction, including integration of predictions of synthetic conditions with new structure predictions, widespread in situ studies to obtain a panoramic view of the reaction mechanism, and the creation of a synthetic database to properly document all synthetic efforts, including the unsuccessful ones.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Bandgap prediction by deep learning in configurationally hybridized graphene and boron nitride

Abstract It is well-known that the atomic-scale and nano-scale configuration of dopants can play a crucial role in determining the electronic properties of materials. However, predicting such effects is challenging due to the large range of atomic configurations that are possible. Here, we present a case study of how deep learning algorithms can enable bandgap prediction in hybridized boron–nitrogen graphene with arbitrary supercell configurations. A material descriptor that enables correlation of structure and bandgap was developed for convolutional neural networks. Bandgaps calculated by ab initio calculations, and corresponding structures, were used as training datasets. The trained networks were then used to predict bandgaps of systems with various configurations. For 4 × 4 and 5 × 5 supercells they accurately predict bandgaps, with a R 2 of >90% and root-mean-square error of ~0.1 eV. The transfer learning was performed by leveraging data generated from small supercells to improve the prediction accuracy for 6 × 6 supercells. This work will pave a route to future investigation of configurationally hybridized graphene and other 2D materials. Moreover, given the ubiquitous existence of configurations in materials, this work may stimulate interest in applying deep learning algorithms for the configurational design of materials across different length scales.

Dong, Yuan↗
DOE OSTI2021

Machine Learning Augmented Predictive and Generative Model for Rupture Life in Ferritic and Austenitic Steels

The Larson-Miller parameter (LMP) offers an efficient and fast scheme to estimate the creep rupture life of alloy materials for high temperature applications. However, owing to poor generalizability and dependence on the constant C, which is typically not known a-priori, estimations using the Larson-Miller parameter often result in suboptimal performance for a wide range of materials. At best it is useful in comparing alloys of similar composition. In this work, three machine learning (ML) schemes were developed for rupture life prediction for 9-12% Cr ferritic-martensitic steels and austenitic stainless steels, i.e., a hierarchical model to parameterize LMP using the LMP constant C to compute rupture life, a hierarchical model to parameterize both C and LMP to compute rupture life, and a direct prediction of rupture life. Specifically, we show that the third scheme, using a gradient boosting algorithm, can be used to train ML models for very accurate prediction of rupture life in a variety of alloys (Pear-son Correlation Coefficient > 0.9 for 9-12% Cr and > 0.8 for austenitic stainless steels). In addition, the Shapley value was used to quantify feature importance, making the model interpretable by identifying the effect of various features on the model performance. Furthermore, a variational autoencoder-based generative model was built by conditioning on the experimental dataset to sample hypothetical synthetic candidate alloys from the learnt joint distribution not existing in both 9-12 % Cr ferritic-martensitic steel and austenitic stainless steel datasets. Finally, based on the predictive and generative model, a reinforcement learning strategy has been proposed to guide experimentalists into designing better heat resistant alloys.

Mamun, Md Osman G.↗
DOE OSTI2025

PQML: Enabling the Predictive Reproducibility on NISQ Machines for Quantum ML Applications

Quantum computing represents a groundbreaking approach to high-performance computing. In recent years, quantum computers have progressed from single-qubit processors to systems boasting over 400 qubits. The presence of such a large number of qubits offers significant advantages, including enhanced computational speed—a capability beyond classical computing methods. However, the current stage of quantum computing is referred to as the noisy intermediate-scale quantum (NISQ) era. The existence of noise in this era presents challenges in testing quantum computing applications, leading to considerable variance in application results. Furthermore, the diverse noise characteristics observed across different machines exacerbate this issue, complicating the selection of the appropriate machine for application execution. In response to these challenges, we introduce our Predictive Quantum Machine Learning (PQML) tool. This tool is designed to predict outcomes when executing identical quantum machine learning applications—specifically, a critical suite of variational quantum algorithms—across various quantum computers during the NISQ era. This effort relies on data collected over a 12-month period. To the best of our knowledge, this study represents the first attempt to ensure reproducibility across quantum computers for complex circuits. Additionally, we have developed a model capable of forecasting the accuracy of quantum computers for variational quantum algorithms, with a particular emphasis on quantum machine learning as a case study.

Senapati, Priyabrata [Kent State University]↗