SEARCH · Engineering Papers

Results for “performance 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 1,477 records · Page 82

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Towards a robust out-of-the-box neural network model for genomic data

The accurate prediction of biological features from genomic data is paramount for precision medicine and sustainable agriculture. For decades, neural network models have been widely popular in fields like computer vision, astrophysics and targeted marketing given their prediction accuracy and their robust performance under big data settings. Yet neural network models have not made a successful transition into the medical and biological world due to the ubiquitous characteristics of biological data such as modest sample sizes, sparsity, and extreme heterogeneity. Here, we investigate the robustness, generalization potential and prediction accuracy of widely used convolutional neural network and natural language processing models with a variety of heterogeneous genomic datasets. Mainly, recurrent neural network models outperform convolutional neural network models in terms of prediction accuracy, overfitting and transferability across the datasets under study. While the perspective of a robust out-of-the-box neural network model is out of reach, we identify certain model characteristics that translate well across datasets and could serve as a baseline model for translational researchers.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2022

Predicted Impacts of Pt and Ionomer Distributions on Low-Pt-Loaded PEMFC Performance

Low-cost, high performance proton exchange membrane fuel cells (PEMFCs) have been difficult to develop due to limited understanding of coupled processes in the cathode catalyst layer (CCL). Low-Pt-loaded PEMFCs suffer losses beyond those predicted solely due to reduced catalyst area. Although consensus links these losses to thin ionomer films in the CCL, a precise mechanistic explanation remains elusive. In this publication, we present a physically based PEMFC model with novel structure-property relationships for thin-film Nafion, validated against PEMFC data with low Pt loading. Results suggest that flooding exacerbates kinetic limitations in low-loaded PEMFCs, shifting the Faradaic current distribution. As current density increases, protons travel further into the CCL, resulting in higher Ohmic overpotentials. We also present a parametric study of CCL design parameters. We find that graded Pt and ionomer loadings reduce Ohmic losses and flooding, but individually do not provide significant improvements. However, a dual-graded CCL (i.e., graded Pt and ionomer) is predicted to significantly improve the maximum power density and limiting current compared to uniformly loaded CCLs. This work highlights the importance of accurate transport parameters for thin-film Nafion and provides a pathway to low-cost PEMFCs via precise control of CCL microstructures.

08 HYDROGEN↗

DOE OSTI2022

Predicting Rupture Pressure and Leak Rate of Alloy 600 Tubes with ODSCC and Comparing with Test Results (Technical Letter Report)

This report summarizes structural integrity work performed for the U.S. NRC that predicts the rupture pressure and leak rate of flawed steam generator (SG) tubes. The work includes pressure testing of straight tubes with lab-induced outer diameter stress corrosion cracks (ODSCC). Existing models for predicting rupture pressure and leak rate of straight tubes with well-defined, rectangular, electro-discharge-machined (EDM) flaws were modified and used to predict the measured rupture pressures and leak rates from straight tubes with ODSCC. Models previously developed for EDM notches in straight tubes were modified for ODSCC by using the equivalent rectangular crack (ERC) method.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2020

A Time Series Sustainability Assessment of a Partial Energy Portfolio Transition

Energy portfolios are overwhelmingly dependent on fossil fuel resources that perpetuate the consequences associated with climate change. Therefore, it is imperative to transition to more renewable alternatives to limit further harm to the environment. This study presents a univariate time series prediction model that evaluates sustainability outcomes of partial energy transitions. Future electricity generation at the state-level is predicted using exponential smoothing and autoregressive integrated moving average (ARIMA). The best prediction results are then used as an input for a sustainability assessment of a proposed transition by calculating carbon, water, land, and cost footprints. Missouri, USA was selected as a model testbed due to its dependence on coal. Of the time series methods, ARIMA exhibited the best performance and was used to predict annual electricity generation over a 10-year period. The proposed transition consisted of a one-percent annual decrease of coal’s portfolio share to be replaced with an equal share of solar and wind supply. The sustainability outcomes of the transition demonstrate decreases in carbon and water footprints but increases in land and cost footprints. Decision makers can use the results presented here to better inform strategic provisioning of critical resources in the context of proposed energy transitions.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

DOE OSTI2019

Intrinsic Exciton Photophysics of PbS Quantum Dots Revealed by Low-Temperature Single Nanocrystal Spectroscopy

With a tunable size-dependent photoluminescence (PL) over a wide infrared wavelength range, lead chalcogenide quantum dots (QDs) have attracted significant scientific and technological interest. Nevertheless, the investigation of intrinsic exciton photophysics at the single-QD level has remained a challenge. Herein, we present a comprehensive study of PL properties for the individual core/shell PbS/CdS QDs emissive near 1.0 eV. In contrast to the sub-meV spectral line widths observed for II/VI QDs, PbS/CdS QDs are predicted to possess broad homogeneous line widths. Performing spectroscopy at cryogenic (4 K) temperatures, we provide direct evidence confirming theoretical predictions, showing that intrinsic line widths for PbS/CdS QDs are in the range of 8–25 meV, with an average of 16.4 meV. In addition, low-temperature, single-QD spectroscopy reveals a broad low-energy side emission attributable to optical as well as localized acoustic phonon-assisted transitions. By tracking single QDs from 4 to 250 K, we were able to probe temperature-dependent evolutions of emission energy, line width, and line shape. Lastly, polarization-resolved PL imaging showed that PbS/CdS QDs are characterized by a 3D emission dipole, in contrast with the 2D dipole observed for CdSe QDs.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

DOE OSTI2024

Machine-Learned Force Field Modeling of Metal Organic Frameworks for CO2 Direct Air Capture

To cope with legacy greenhouse gas emissions and to achieve net-zero emissions by 2050, the U.S. Department of Energy (DOE) is funding efforts to develop direct air capture (DAC), a method for removing CO2 directly from air. Metal organic frameworks (MOFs) have been well studied as DAC sorbent materials due to their tunable structural and compositional properties. Thermodynamic simulations using force fields are often used to provide predictions of a material’s performance in many separations. However, these force fields often make assumptions about bonds and the physics of the adsorption process. A new class of force fields called machine-learned force fields (MLFFs) use machine learning to form quantitative relationships between a material’s chemical structure and the forces and energies predicted by more accurate quantum mechanical calculations, such as dispersion-corrected density functional theory (DFT). In this work, models were developed to achieve DFT-level accuracy for the forces and energies associated with MOF flexibility and CO2 adsorption using MLFFs. These methods were parametrized based on thousands of DFT calculations of CO2 in flexible MOFs and used to predict MOF structural properties as well as CO2 adsorption in several MOFs.

Findley, John↗

DOE OSTI2021

Lessons Learned in Employing Data Analytics to Predict Oxidation Kinetics and Spallation Behavior of High-Temperature NiCr-Based Alloys

Machine learning (ML) can offer many advantages in predicting material properties over traditional materials development methods based solely on limited experimental investigations or physical-based simulations with the capability to reduce development cost, risk, and time. However, so far, limited efforts have been made to predict alloy oxidation kinetics and spallation behavior via ML due to the lack of consistently measured and sufficient experimental data and the inherent complexity in oxidation behavior of multicomponent high-temperature alloys. A previous study reported the ability of ML to predict oxidation kinetics of NiCr-based alloys as a function of alloy composition and operating conditions. Here, the performance of a ML model in predicting rate constants and spallation probability was evaluated in light of the roles of the data distribution of the experimental dataset (data analytics), the alloy composition, the exposure environment and the chosen oxidation approach to extracting kinetic values from the measured mass changes (but using either a simple parabolic law or a statistical cyclic oxidation model). Potential strategies to improve the predictions and enhance the extrapolative capability of the previously trained model will be discussed.

36 MATERIALS SCIENCE↗

DOE OSTI2025

In Silico Design of Methyl-Driven Overhauser Dynamic Nuclear Polarization Agents

Overhauser effect (OE) dynamic nuclear polarization (DNP) has drawn attention owing to its enhanced performance at ultrahigh magnetic fields. The lack of design principles has, nevertheless, limited the development of OE polarizing agents when compared to those used for cross-effect DNP. Here, we measured the 19 F OE DNP performance of a series of CF 3 -functionalized Blatter-type radicals. Using density functional theory calculations, we accurately predict the methyl-driven OE performance, paving the way for computer-aided design of optimized OE DNP polarizing agents.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2023

Improved Modeling of Waste Isolation Pilot Plant Disposal Room Porosity

Disposal rooms at the Waste Isolation Pilot Plant (WIPP) contain waste and gas, and their porosity evolves over time. This report presents several improvements to the disposal room porosity model and presents new porosity predictions for use in future WIPP Performance Assessment activities. The improvements pertain to three sub-models: the geomechanical model, the waste compaction model, and the gas pressurization model. The impacts of each major improvement were quantified and the new porosity predictions were shown to be both mesh and domain size converged. Also, sensitivity studies on the disposal room horizon, clay seam friction coefficients, and homogenized waste representation were performed to support assumptions in the disposal room porosity model. To compare the legacy and new porosity predictions, the simulation results were plotted as a response surface, where gas pressure and time are inputs and porosity is the output. The new porosity response surface is insensitive to pressures beneath lithostatic pressure and highly sensitive to pressures above lithostatic pressure. The legacy porosity response surface, on the other hand, has moderate porosity gradients over all pressures. The new porosity response surface has a stronger scientific foundation than the legacy surface and may now be used for Compliance Decision Analyses.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

DOE OSTI2024

Active learning of ternary alloy structures and energies

Abstract Machine learning models with uncertainty quantification have recently emerged as attractive tools to accelerate the navigation of catalyst design spaces in a data-efficient manner. Here, we combine active learning with a dropout graph convolutional network (dGCN) as a surrogate model to explore the complex materials space of high-entropy alloys (HEAs). We train the dGCN on the formation energies of disordered binary alloy structures in the Pd-Pt-Sn ternary alloy system and improve predictions on ternary structures by performing reduced optimization of the formation free energy, the target property that determines HEA stability, over ensembles of ternary structures constructed based on two coordinate systems: (a) a physics-informed ternary composition space, and (b) data-driven coordinates discovered by the Diffusion Maps manifold learning scheme. Both reduced optimization techniques improve predictions of the formation free energy in the ternary alloy space with a significantly reduced number of DFT calculations compared to a high-fidelity model. The physics-based scheme converges to the target property in a manner akin to a depth-first strategy, whereas the data-driven scheme appears more akin to a breadth-first approach. Both sampling schemes, coupled with our acquisition function, successfully exploit a database of DFT-calculated binary alloy structures and energies, augmented with a relatively small number of ternary alloy calculations, to identify stable ternary HEA compositions and structures. This generalized framework can be extended to incorporate more complex bulk and surface structural motifs, and the results demonstrate that significant dimensionality reduction is possible in thermodynamic sampling problems when suitable active learning schemes are employed.

Chemistry↗

DOE OSTI2022

The study of C-13(a, n)O-16 reactions in the Majorana Demonstrator calibration data

Neutron captures and delayed decays of reaction products are common sources of backgrounds in ultra-rare event searches. In this work, we studied 13C(a, n)1640 O reactions induced by a-particles emitted within the calibration sources of the Majorana Demonstrator. These sources are thorium-based calibration standards enclosed in carbon-rich materials. The reaction rate was estimated by using the 6129 keV ?-rays emitted from the excited 1643 O states that are populated when the incoming a-particles have enough energy. Thanks to the excellent energy performance of the Demonstrator’s germanium detectors, these characteristic photons can be clearly observed in the calibration data. Facilitated by Geant4 simulations, a comparison between the observed 6129 keV photon rates and predictions by a TALYS-based software was performed. The measurements and predictions were found to be consistent with each other, albeit with large statistical uncertainties. This agreement provides supports for further background predictions for (a, n)-reactions in future double-beta decay search efforts.

6 ≤ A ≤ 19↗

DOE OSTI2026

Continual Learning for Production-Level Machine Learning in Particle Accelerators

Particle accelerators operate in complex environments where data distribution can change dynamically, leading to data drifts that significantly challenge Machine Learning (ML) models. These non-stationary conditions often cause ML models to deteriorate in performance, making it difficult to maintain reliable predictions in operation. The primary sources of data drifts are changes in accelerator settings and changes in equipment performance which cannot be measured directly. To bridge this gap between ML development and long-term deployment in operational settings, we identify key areas within particle accelerators where continual learning can help mitigate drift-induced performance degradation. We will provide a practical guide on selecting the appropriate method given resource constraints and desired stability plasticity trade offs. As a concrete example, we will present a real-world use case for anomaly detection to predict errant beams at the Spallation Neutron Source accelerator, where continual learning has been employed to demonstrate stable performance on drifting data streams. We will present practical challenges, lessons learned, and the results from the deployed ML model.

Rajput, Kishansingh [Thomas Jefferson National Acc↗

DOE OSTI2025

Study of CORC Conductors With Respect to Individual Tape Properties

CEA has been studying the advantages of a conductor based on an assembly of CORC cables in the high field zone of a hybrid Central Solenoid (CS) magnet for EU-DEMO. To this end, the detailed study of geometrical and electrical parameters of a CORC structure are studied in order to evaluate the cable’s electrical performance as well as to determine a number of important parameters (crossing points, contact surface etc…) as function of the cable structure. The paper first presents these geometrical and performance analyses. It will then try to introduce smeared models in order to reduce the cable’s performance to 1D tape scaling law using effective parameters. That reduction is of importance for use in thermohydraulic models. Finally, the paper will present the case study of a particular CORC conductor that is being procured and is foreseen to be tested in SULTAN in 2025. The paper also will try to give some predictive estimate of the cable performance and identify some of the unknowns related to current redistribution and joint resistance.

CORC↗

DOE OSTI2020

An a priori evaluation of a principal component and artificial neural network based combustion model in diesel engine conditions

A principal component analysis (PCA) and artificial neural network (ANN) based chemistry tabulation approach is presented. ANNs are used to map the thermochemical state onto a low-dimensional manifold consisting of five control variables that have been identified using PCA. Three canonical configurations are considered to train the PCA-ANN model: a series of homogeneous reactors, a nonpremixed flamelet, and a two-dimensional lifted flame. The performance of the model in predicting the thermochemical manifold of a spatially-developing turbulent jet flame in diesel engine thermochemical conditions is a priori evaluated using direct numerical simulation (DNS) data. The PCA-ANN approach is compared with a conventional tabulation approach (tabulation using ad hoc defined control variables and linear interpolation). The PCA-ANN model provides higher accuracy and requires several orders of magnitude less memory. Here, these observations indicate that the PCA-ANN model is superior for chemistry tabulation, especially for modelling complex chemistries that present multiple combustion modes as observed in diesel combustion. The performance of the PCA-ANN model is then compared to the optimal estimator, i.e. the conditional mean from the DNS. The results indicate that the PCA-ANN model gives high prediction accuracy, comparable to the optimal estimator, especially for major species and the thermophysical properties. Higher errors are observed for the minor species and reaction rate predictions when compared to the optimal estimator. It is shown that the prediction of minor species and reaction rates can be improved by using training data that exhibits a variation of parameters as observed in the turbulent flame. The output of the ANN is analysed to assess mass conservation. It is observed that the ANN incurs a mean absolute error of 0.05% in mass conservation. Furthermore, it is demonstrated that this error can be reduced by modifying the cost function of the ANN to penalise for deviation from mass conservation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2019

Accelerated Search for BaTiO 3 -Based Ceramics with Large Energy Storage at Low Fields Using Machine Learning and Experimental Design

The problem that is considered is that of maximizing the energy storage density of Pb-free BaTiO3-based dielectrics at low electric fields. It is demonstrated that how varying the size of the combinatorial search space influences the efficiency of material discovery by comparing the performance of two machine learning based approaches where different levels of physical insights are involved. It is started with physics intuition to provide guiding principles to find better performers lying in the crossover region in the composition–temperature phase diagram between the ferroelectric phase and relaxor ferroelectric phase. Such an approach is limiting for multidopant solid solutions and motivates the use of two data-driven machine learning and design strategies with a feedback loop to experiments. Strategy I considers learning and property prediction on all the compounds, and strategy II learns to preselect compounds in the crossover region on which prediction is carried out. By performing only two active learning loops via strategy II, the compound (Ba0.86Ca0.14)(Ti0.79Zr0.11Hf0.10)O3 is synthesized with the largest energy storage density ≈73 mJ cm-3 at a field of 20 kV cm-1 , and an insight into the relative performance of the strategies using varying levels of knowledge is provided.

36 MATERIALS SCIENCE↗

DOE OSTI2021

Machine learning application to single channel design of molten salt reactor

This study proposes a robust approach to quickly design a nuclear reactor core and explores the best performing machine learning (ML) technique for predicting feature parameters of the core. Here we implemented the approach into a hypothetical channel of molten salt reactors to demonstrate the applicability of the method. We prepared a Python tool, named Plankton, which couples to a reactor physics code and an optimization tool, and imports ML methods. The tool performs three consecutive phases: reactor database generation, machine learning application, and design optimization. We identified the extra trees method as the best performing estimator. With the estimator, we found nine optimum designs in total, one for each fuel-salt pair, and estimated all the performance metrics of the designs with a <5% prediction error compared to their actual values. U-Pu-NaCl fuel-salt gave promising results with the highest conversion ratio, the most negative feedback coefficient, and the lowest fast flux.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

DOE OSTI2021

Predictive Battery Lifetime Modeling at NREL [Slides]

Battery lifetime models are used to extrapolate data from accelerated aging tests to simulate degradation in real-world applications such as electric vehicles and battery energy storage systems. Methods developed at NREL utilize both expert domain-knowledge and machine-learning to identify models, using statistical methods such as cross-validation and bootstrap resampling to interrogate model performance and quantify uncertainty. These models can be utilized in systems level simulations to predict battery performance or technoeconomic models to estimate the lifetime cost of battery systems.

25 ENERGY STORAGE↗

DOE OSTI2021

Fracture Sustainability in Enhanced Geothermal Systems: Experimental and Modeling Constraints

Enhanced geothermal systems (EGS) offer the potential for a much larger energy source than conventional hydrothermal systems. Hot, low-permeability rocks are prevalent at depth around the world, but the challenge of extracting thermal energy depends on the ability to create and sustain open fracture networks. Laboratory experiments were conducted using a suite of selected rock cores (granite, metasediment, rhyolite ash-flow tuff, and silicified rhyolitic tuff) at relevant pressures (uniaxial loading up to 20.7 MPa and fluid pressures up to 10.3 MPa) and temperatures (150–250 °C) to evaluate the potential impacts of circulating fluids through fractured rock by monitoring changes in fracture aperture, mineralogy, permeability, and fluid chemistry. Because a fluid in disequilibrium with the rocks (deionized water) was used for these experiments, there was net dissolution of the rock sample: this increased with increasing temperature and experiment duration. Thermal-hydrological-mechanical-chemical (THMC) modeling simulations were performed for the rhyolite ash-flow tuff experiment to test the ability to predict the observed changes. These simulations were performed in two steps: a thermal-hydrological-mechanical (THM) simulation to evaluate the effects of compression of the fracture, and a thermal-hydrological-chemical (THC) simulation to evaluate the effects of hydrothermal reactions on the fracture mineralogy, porosity, and permeability. Furthermore, these experiments and simulations point out how differences in rock mineralogy, fluid chemistry, and geomechanical properties influence how long asperity-propped fracture apertures may be sustained. Such core-scale experiments and simulations can be used to predict EGS reservoir behavior on the field scale.

15 GEOTHERMAL ENERGY↗