Search NASA⌕ Search

SEARCH · Search NASA

Results for “Predictive modeling”

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 1,099 records · Page 61

Assessment of an atmospheric transport model for annual inverse estimates of California greenhouse gas emissions

Abstract Atmospheric inverse estimates of gas emissions depend on transport model predictions, hence driving a need to assess uncertainties in the transport model. In this study we assess the uncertainty in WRF‐STILT (Weather Research and Forecasting and Stochastic Time‐Inverted Lagrangian Transport) model predictions using a combination of meteorological and carbon monoxide (CO) measurements. WRF configurations were selected to minimize meteorological biases using meteorological measurements of winds and boundary layer depths from surface stations and radar wind profiler sites across California. We compare model predictions with CO measurements from four tower sites in California from June 2013 through May 2014 to assess the seasonal biases and random errors in predicted CO mixing ratios. In general, the seasonal mean biases in boundary layer wind speed (< ~ 0.5 m/s), direction (< ~ 15°), and boundary layer height (< ~ 200 m) were small. However, random errors were large (~1.5–3.0 m/s for wind speed, ~ 40–60° for wind direction, and ~ 300–500 m for boundary layer height). Regression analysis of predicted and measured CO yielded near‐unity slopes (i.e., within 1.0 ± 0.20) for the majority of sites and seasons, though a subset of sites and seasons exhibit larger (~30%) uncertainty, particularly when weak winds combined with complex terrain in the South Central Valley of California. Looking across sites and seasons, these results suggest that WRF‐STILT simulations are sufficient to estimate emissions of CO to up to 15% on annual time scales across California.

Bagley, Justin E.↗

A Mathematical Model for Prediction of Carbon Concentration During RH Refining Process

A mathematical model was developed to predict the carbon concentration during RH refining process. Three reaction sites such as the free surface, inner sites of the molten steel in vacuum chamber, and the bubble surface were considered. The decarburization at inner sites of the molten steel in vacuum chamber was dominant at the initial stage, and then subsequently, it was dominated by the free surface and bubble surface. The decarburization rate constant was also evaluated.

36 MATERIALS SCIENCE↗

The importance of atmospheric turbulence and stability in machine-learning models of wind farm power production

Machine learning is frequently applied in the wind energy industry to build statistical models of wind farm power production using atmospheric data as input. In the field of wind power forecasting, in particular, there has been substantial research into finding the best-performing learning algorithms that improve model predictions. Overlooked in the literature, however, is the influence of atmospheric turbulence and stability measurements in improving model predictions. It has been well-established through observations and physical models that these effects can have considerable influence on wind farm power production; yet consideration of these effects in statistical models is almost entirely absent from the literature. In this work, we examine the impact of atmospheric turbulence and stability inputs on statistical model predictions of wind farm power output. Hourly observations from a wind farm in the Pacific Northwest United States located in very complex terrain are used. Five common learning algorithms and nine atmospheric variables are considered, five of which represent some measure of turbulence or stability. Here, we find a considerable improvement in hourly power predictions when some measure of turbulence or stability is included in the model. In particular, turbulent kinetic energy was found to be the most important variable apart from wind speed and more important than wind direction, pressure, and temperature. By contrast, the choice of learning algorithm is shown to be relatively less important in improving predictions. Based on this work, we recommend that turbulence and stability variables become standard inputs into statistical models of wind farm power production.

17 WIND ENERGY↗

A Time-Varying Model for Predicting Formaldehyde Emission Rates in Homes

Recent studies have succeeded in relating emissions of various volatile organic compounds to material mass diffusion transfer using detailed empirical characteristics of each of the individual emitting materials. While significant, the resulting models are often scenario specific and/or require a host of individual component parameters to estimate emission rates. This study developed an approach to estimate aggregated emissions rates based on a wide number of field measurements. We used a multi-parameter regression model based on previous mass transfer models to predict formaldehyde emission rate for a whole dwelling using field-measured, time-resolved formaldehyde concentrations, air exchange rates, and indoor environmental parameters in 63 California single-family houses built between 2011 and 2017. The resulting model provides time-varying formaldehyde emission rates, normalized by floor area, for each study home, assuming a well-mixed mass balance transport model of the home, and a well-mixed layer transport model of indoor surfaces. The surface layer model asserts an equilibrium concentration within the surface layer of the emitted materials that is a function of temperature and RH; the dwelling ventilation rate serves as a surrogate for indoor concentration. We also developed a more generic emission model that is suitable for broad prediction of emission for a population of buildings. This model is also based on measurements aggregated from 27 homes from the same study. We showed that errors in predicting household formaldehyde concentrations using this approach were substantially less than those using a traditional constant emission rate model, despite requiring less unique building information.

54 ENVIRONMENTAL SCIENCES↗

Measurement and analysis of release isentropes using proton radiography (u)

Off-Hugoniot measurements are needed to further develop predictive models that accurately describe the behavior of metals undergoing phase transitions. Predictive modeling of phase transitions is essential for LANL to meet its programmatic objectives. Understanding the dynamic evolution of density as a function of time during the release process is important to developing high fidelity equation of state models. This is particularly true for metals that have a degree of complexity, such as a solid-solid phase transition. The equations-of-state (EOS) for metals with complexity are more difficult to measure and to model, and states far away from where measurements are easily made can be poorly known. Accurate density measurements can provide us with additional fundamental information that can be used to further constrain the equation of state for a material. Currently release isentrope information is obtained from shock experiments at a sample window interface using optical velocimetry. This data is highly convoluted due to wave interactions between the sample window-interface making it difficult to infer physical processes happening within the material. Proton radiography has the ability to probe the release waves in-situ before these wave interactions can take place. Since multiple radiographs are obtained in each experiment, pRad provides the unique capability of being able to measure both density and wave evolution within the material. The measurement of release wave densities, however, presents new challenges for pRad since the release wave of a shocked transition is not a step function but instead a ramped wave. Previous pRad experiments have generally measured density jumps over step transitions; therefore, new analysis techniques will have to be developed to measure the ramped density change of a release isentrope. Once developed, these new analysis techniques can also be used for experiments that involve ramped compression. This kind of compression is being used to measure states close to an isentrope. The ability analyze ramp wave data is an essential component to further the use of proton radiography to study any off-Hugoniot behavior. Wave speeds in materials depend upon the speed of sound which in turn is a function of pressure in the material. At higher pressures the sound speed is faster; at lower pressures the sound speed is slower; this is what causes shock waves to form. This same effect causes the shape of release waves to change with time. The wave will be running faster at the top and slower at the bottom, its slope steeper at the top and less at the bottom due to pressure changes. These are unsteady waves. WONDY simulations of these wave shape characteristics are shown in Fig. 1 for a Cu symmetric impact. We will need to look at how the shape of the wave changes over time and affects the measurement of the density as a function of time. Additionally we need to quantify the uncertainties in the spatial and density measurements as a function of ramp rate and shock strength. We will look at aluminum or copper to validate the analysis techniques that will need to be developed. Finally, we will perform shock loaded phase transition experiments on iron, zirconium, and tin using the developed analysis techniques for incorporation into predictive models. The iron (and possibly tin) experiments will introduce a new degree of complexity because of the existence of a rarefaction shock in the release wave.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Comparison of structurally diverse simulation models for prediction of epidemic outcomes caused by a long-distance dispersed pathogen

Long-distance dispersal (LDD) pathogens pose substantial challenges for epidemic control due to their ability to generate new infection foci at great distances. While various modeling approaches have been developed to understand and manage such outbreaks, little work has compared how models of different structures behave under shared conditions. Here, in this study, we compare four structurally distinct epidemiological models — EPIMUL, GEMF, PoPS, and Warwick — each adapted to simulate the spread of wheat stripe rust (WSR), a wind-dispersed LDD pathogen, under identical epidemiological parameters and dispersal kernel. Using data from a controlled field experiment, we evaluate the ability of each model to replicate disease prevalence under nine intervention scenarios that vary in timing and culling area. While the models differ substantially in design — ranging from spatial grid-based to network-based and raster-based frameworks — the shared dispersal kernel allowed for close alignment in their predictions. All models accurately captured general epidemic trends, particularly the strong effect of early intervention on disease suppression. We qualitatively compared their behavioral responses across scenarios and also evaluated an ensemble prediction by averaging across model outputs. Our findings highlight how integrating shared epidemiological components into distinct modeling frameworks can improve consistency and accuracy, while reinforcing the importance of early culling in managing LDD pathogen outbreaks.

Dispersal kernel↗

Comparative hazard analysis and toxicological modeling of diverse nanomaterials using the embryonic zebrafish (EZ) metric of toxicity

The integration of rapid assays, large data sets, informatics and modeling can overcome current barriers in understanding nanomaterial structure-toxicity relationships by providing a weight-of-the-evidence mechanism to generate hazard rankings for nanomaterials. Here we present the use of a rapid, low-cost assay to perform screening-level toxicity evaluations of nanomaterials in vivo. Calculated EZ Metric scores, a combined measure of morbidity and mortality, were established at realistic exposure levels and used to develop a predictive model of nanomaterial toxicity. Hazard ranking and clustering analysis of 68 diverse nanomaterials revealed distinct patterns of toxicity related to both core composition and outermost surface chemistry of nanomaterials. The resulting clusters guided the development of a predictive model of gold nanoparticle toxicity to embryonic zebrafish. In addition, our findings suggest that risk assessments based on the size and core composition of nanomaterials alone may be wholly inappropriate, especially when considering complex engineered nanomaterials. These findings reveal the need to expeditiously increase the availability of quantitative measures of nanomaterial hazard and broaden the sharing of that data and knowledge to support predictive modeling. In addition, research should continue to focus on methodologies for developing predictive models of nanomaterial hazard based on sub-lethal responses to low dose exposures.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Simulation of complex glazing products; from optical data measurements to model based predictive controls

Complex glazing systems such as venetian blinds, fritted glass and woven shades require more detailed optical and thermal input data for their components than specular non light-redirecting glazing systems. Various methods for measuring these data sets are described in this paper. These data sets are used in multiple simulation tools to model the thermal and optical properties of complex glazing systems. The output from these tools can be used to generate simplified rating values or as an input to other simulation tools such as whole building annual energy programs, or lighting analysis tools. I also describe some of the challenges of creating a rating system for these products and which factors affect this rating. A potential future direction of simulation and building operations is model based predictive controls, where detailed computer models are run in real-time, receiving data for an actual building and providing control input to building elements such as shades.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Predictive AdS/QCD model for mass spectra of mesons with three flavors

The predictive soft-wall AdS/QCD model with a modified 5D metric at the infrared region is constructed to obtain a nontrivial dilaton solution for three flavor quarks u, d and s. Such a model is shown to incorporate both the chiral symmetry breaking and linear confinement. After considering some high-order terms including the U(1){sub L}xU(1){sub R} chiral symmetry breaking term, we find that the resulting predictions for the SU(3) octet and singlet resonance states of pseudoscalar, scalar, vector and axial-vector mesons agree well with the experimentally confirmed resonance states. Contributions from the instanton effects given by the determinant term are also discussed. It is observed that the chiral symmetry breaking phenomena of SU(3){sub L}xSU(3){sub R} and U(1){sub L}xU(1){sub R} can be well described in this model, while the SU(3) flavor symmetry breaking effect due to quark mass difference in the source term is not enough to explain all of the current experimental data.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

TwoFold: Highly accurate structure and affinity prediction for protein-ligand complexes from sequences

We describe our development of ab initio protein-ligand binding pose prediction models based on transformers and binding affinity prediction models based on the neural tangent kernel (NTK). Folding both protein and ligand, the TwoFold models achieve efficient and quality predictions matching state-of-the-art implementations while additionally reconstructing protein structures. In conclusion, solving NTK models points to a new use case for highly optimized linear solver benchmarking codes on HPC.

60 APPLIED LIFE SCIENCES↗

Machine learning models for rat multigeneration reproductive toxicity prediction

Reproductive toxicity is one of the prominent endpoints in the risk assessment of environmental and industrial chemicals. Due to the complexity of the reproductive system, traditional reproductive toxicity testing in animals, especially guideline multigeneration reproductive toxicity studies, take a long time and are expensive. Therefore, machine learning, as a promising alternative approach, should be considered when evaluating the reproductive toxicity of chemicals. We curated rat multigeneration reproductive toxicity testing data of 275 chemicals from ToxRefDB (Toxicity Reference Database) and developed predictive models using seven machine learning algorithms (decision tree, decision forest, random forest, k-nearest neighbors, support vector machine, linear discriminant analysis, and logistic regression). A consensus model was built based on the seven individual models. An external validation set was curated from the COSMOS database and the literature. The performances of individual and consensus models were evaluated using 500 iterations of 5-fold cross-validations and the external validation data set. The balanced accuracy of the models ranged from 58% to 65% in the 5-fold cross-validations and 45%–61% in the external validations. Prediction confidence analysis was conducted to provide additional information for more appropriate applications of the developed models. The impact of our findings is in increasing confidence in machine learning models. We demonstrate the importance of using consensus models for harnessing the benefits of multiple machine learning models (i.e., using redundant systems to check validity of outcomes). While we continue to build upon the models to better characterize weak toxicants, there is current utility in saving resources by being able to screen out strong reproductive toxicants before investing in vivo testing. The modeling approach (machine learning models) is offered for assessing the rat multigeneration reproductive toxicity of chemicals. Our results suggest that machine learning may be a promising alternative approach to evaluate the potential reproductive toxicity of chemicals.

consensus model↗

A one-dimensional Fickian model to predict the Ga depth profiles in three-stage Cu(In,Ga)Se{sub 2}

We present a one-dimensional Fickian model that predicts the formation of a double Ga gradient during the fabrication of Cu(In,Ga)Se{sub 2} thin films by three-stage thermal co-evaporation. The model is based on chemical reaction equations, structural data, and effective Ga diffusivities. In the model, the Cu(In,Ga)Se{sub 2} surface is depleted from Ga during the deposition of Cu-Se in the second deposition stage, leading to an accumulation of Ga near the back contact. During the third deposition stage, where In-Ga-Se is deposited at the surface, the atomic fluxes within the growing layer are inverted. This results in the formation of a double Ga gradient within the Cu(In,Ga)Se{sub 2} layer and reproduces experimentally observed Ga distributions. The final shape of the Ga depth profile strongly depends on the temperatures, times and deposition rates used. The model is used to evaluate possible paths to flatten the marked Ga depth profile that is obtained when depositing at low substrate temperatures. We conclude that inserting Ga during the second deposition stage is an effective way to achieve this.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Comparison of SUSY spectra generators for natural SUSY and string landscape predictions

Models of natural supersymmetry give rise to a weak scale m weak ~ m W,Z,h ~ 100 GeV without any (implausible) finetuning of independent contributions to the weak scale. These models, which exhibit radiatively driven naturalness (RNS), are expected to arise from statistical analysis of the string landscape wherein large soft terms are favored, but subject to a not-too-large value of the derived weak scale in each pocket universe of the greater multiverse. The string landscape picture then predicts, using the Isajet SUSY spectra generator Isasugra, a statistical peak at m h ~ 125 GeV with sparticles generally beyond current LHC search limits. In this paper, we investigate how well these conclusions hold up using other popular spectra generators: SOFTSUSY, SPHENO and SUSPECT (SSS). We built a computer code DEW4SLHA which operates on SUSY Les Houches Accord files to calculate the associated electroweak naturalness measure Δ EW . The SSS generators tend to yield a Higgs mass peak ~125–127 GeV with a superparticle mass spectra rather similar to that generated by Isasugra. In an Appendix, we include loop corrections to Δ EW in a more standard notation.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Aging Behavior of Viton{sup R} O-Ring Seals in the 9975 Shipping Package - 12594

The Savannah River Site (SRS) is storing plutonium (Pu) materials in the K-Area Materials Storage (KAMS) facility. The Pu materials were packaged according to the DOE-STD-3013 standard and shipped to the SRS in Type B 9975 packages. The robust 9975 shipping package was not designed for long-term product storage, but it is a specified part of the storage configuration and the KAMS facility safety basis credits the 9975 design with containment. Within the 9975 package, nested stainless steel containment vessels are closed with dual O-ring seals based on Viton{sup R} GLT or GLT-S fluoro-elastomer. The aging behavior of the O-ring compounds is being studied to provide the facility with advanced notice of nonconformance and to develop life prediction models. A combination of field surveillance, leak testing of surrogate fixtures aged at bounding service temperatures, and accelerated-aging methodologies based on compression stress-relaxation and oxygen consumption analysis is being used to evaluate seal performance. A summary of the surveillance program relative to seal aging behavior is presented. The aging behavior of fluoro-elastomer seals based on Viton{sup R} GLT and GLT-S is being studied to develop life prediction models in support of long-term storage of plutonium materials in the 9975 shipping packages at the Savannah River Site. Field surveillance data in combination with accelerated-aging data suggest a significant lifetime for the seals. Typical storage conditions are not anticipated to challenge the leak-tightness of the seals for many years. Early life prediction models based on compression stress relaxation indicate a seal lifetime of ∼12 years at the maximum service temperature predicted (93 deg. C). Seal lifetimes at lower, more realistic conditions are likely significantly longer. Service life predictions based on CSR data are thus far conservative relative to predictions based on time to leakage failure. Surveillance data on packages examined after 6 years in storage show only minor compression set of the O-rings and no significant degradation. Surveillance and testing will continue as needed to validate and refine life prediction models. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

An Artificial Neural Network Model to Predict the Bainite Plate Thickness of Nanostructured Bainitic Steels Using an Efficient Network-Learning Algorithm

Nanostructured bainitic steel has an extraordinary ultrahigh strength of about 2.0 GPa along with good toughness of 30 MPa m{sup 1/2}. The finer thickness of plate 20-80 nm is largely responsible for achieving such a large hardness of 690-720 HV. In this work, a multilayer perceptron-based artificial neural network (ANN) model has been developed to predict the thickness of bainite plate pertaining to nanostructured bainitic steels. The inputs of the ANN model are, namely, Gibbs free energy for bainitic transformation, austenite strength, transformation temperature for bainite and carbon concentration in the steel. The model prediction revealed that the bainite plate thickness critically depends on austenite strength and Gibbs free energy. From the neural prediction, it is concluded that formation of nanostructured bainitic steel is feasible only if sufficient austenite strength (above 165 MPa) has been achieved. This can be accomplished with a minimum carbon content of 0.5 wt.% in steel and transformation temperature below 300 °C. Higher driving force (Gibbs free energy) below − 1800 J/mol is another prerequisite condition of formation of nanostructured bainite steel. The network-learning architecture has been optimized using the Broyden–Fletcher–Goldfarb–Shanno (BFGS) algorithm to minimize the network training error within eight training cycles. The algorithm facilitates a faster convergence of network training and testing errors.

36 MATERIALS SCIENCE↗

Be 7 solar neutrino measurement with KamLAND

In this article, we report a measurement of the neutrino-electron elastic scattering rate of 862 keV 7 Be solar neutrinos based on a 165.4 kt d exposure of KamLAND. The observed rate is 582 ± 94 (kt d) -1 , which corresponds to an 862-keV 7 Be solar neutrino flux of (3.26 ± 0.52) × 10 9 cm -2 s -1 , assuming a pure electron-flavor flux. Comparing this flux with the standard solar model prediction and further assuming three-flavor mixing, a ν e survival probability of 0.66 ± 0.15 is determined from the KamLAND data. Utilizing a global three-flavor oscillation analysis, we obtain a total 7 Be solar neutrino flux of (5.82 ± 1.02) × 10 9 cm -2 s -1 , which is consistent with the standard solar model predictions.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Allometric Models for Predicting Aboveground Biomass and Carbon Stock of Tropical Perennial C 4 Grasses in Hawaii

Biomass is a promising renewable energy option that provides a more environmentally sustainable alternative to fossil resources by reducing the net flux of greenhouse gasses to the atmosphere. Yet, allometric models that allow the prediction of aboveground biomass (AGB), biomass carbon (C) stock non-destructively have not yet been developed for tropical perennial C 4 grasses currently under consideration as potential bioenergy feedstock in Hawaii and other subtropical and tropical locations. The objectives of this study were to develop optimal allometric relationships and site-specific models to predict AGB, biomass C stock of napiergrass, energycane, and sugarcane under cultivation practices for renewable energy and validate these site-specific models against independent data sets generated from sites with widely different environments. Several allometric models were developed for each species from data at a low elevation field on the island of Maui, Hawaii. A simple power model with stalk diameter (D) was best related to AGB and biomass C stock for napiergrass, energycane, and sugarcane, (R 2 = 0.98, 0.96, and 0.97, respectively). The models were then tested against data collected from independent fields across an environmental gradient. For all crops, the models over-predicted AGB in plants with lower stalk D, but AGB was under-predicted in plants with higher stalk D. The models using stalk D were better for biomass prediction compared to dewlap H (Height from the base cut to most recently exposed leaf dewlap) models, which showed weak validation performance. Although stalk D model performed better, however, the mean square error (MSE)-systematic was ranged from 23 to 43 % of MSE for all crops. A strong relationship between model coefficient and rainfall was existed, although these were irrigated systems; suggesting a simple site-specific coefficient modulator for rainfall to reduce systematic errors in water-limited areas. These allometric equations provide a tool for farmers in the tropics to estimate perennial C4 grass biomass and C stock during decision-making for land management and as an environmental sustainability indicator within a renewable energy system.

54 ENVIRONMENTAL SCIENCES↗