Engineering Papers⌕ Search

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.

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

At least 1,099 records · Page 61
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Assessment of Buffer-IPyC Thermomechanical Debonding Behavior Using New Experiment Strength Data in BISON

TRIstructural ISOtropic (TRISO) single sub-millimeter-diameter nuclear fuel particle consists of a spherical fuel kernel surrounded by four coating layers. During reactor operation, the TRISO coating layers are subjected to irradiation-induced dimensional changes and the associated thermomechanical behavior of each layer. One of the observed behaviors is gap formation between the buffer and IPyC layer due to the porous buffer’s irradiation-induced shrinkage exceeding that of the IPyC layer. To predict TRISO thermomechanical performance, multi-physics models have been built that are continually updated and refined. The BISON code is a finite element-based nuclear fuel performance code that may be used for TRISO particle simulations. One of the recent additions to the BISON code is the ability to model the process of layer debonding. This paper will focus on the simulation results of the improved BISON debonding model that will utilize updated strengths measured from irradiated TRISO fuel particles. The new experimental strength data from micromechanical tests of irradiated TRISO fuel samples were exercised in the BISON simulations and compared to baseline strength data to assess their applicability in the models. This also includes updated buffer-IPyC bond strengths to simulate layer delamination. Based on current experimental observations it is noted that the buffer-IPyC separation occurs not exactly at the junction of these two layers, but more on the side of the buffer layer. This improved modeling approach using experimental strength data to characterize buffer-IPyC debonding and its potential subsequent cracking will be presented in the paper along with comparisons to available experimental observations.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2025

Assessment of buffer-IPyC thermomechanical debonding behavior using new experimental strength data in BISON

TRIstructural ISOtropic (TRISO) fuel is a nuclear fuel commonly used in High Temperature Gas-cooled Reactors (HTGRs). A single sub-millimeter-diameter TRISO fuel particle consists of a spherical fuel kernel surrounded by four coating layers: a low-density pyrocarbon buffer layer, an inner pyrolytic carbon (IPyC) layer, a silicon carbide (SiC) layer, and an outer pyrolytic carbon (OPyC) layer. The kernel is commonly made of UO2 or a mixture of uranium carbide and uranium oxide (UCO). During reactor operation, the TRISO coating layers are subjected to irradiation-induced dimensional changes and the associated thermomechanical behavior of each layer. One of the observed behaviors is gap formation between the buffer and IPyC layer due to the porous buffer’s irradiation-induced shrinkage exceeding that of the IPyC layer. Not all irradiated particles will experience buffer-IPyC gap formation. The debonding may be partial, or it may be nearly total. However, from post-irradiation examination of UCO TRISO fuels irradiated as part of the Advanced Gas Reactor (AGR) Fuel Development and Qualification Program, it was concluded that partial buffer-IPyC debonding was the most common type of buffer-IPyC interaction. To predict TRISO thermomechanical performance, multi-physics models have been built that are being continually updated and refined. The BISON code is a finite element-based nuclear fuel performance code that may be used for 1D, 2D, and 3D TRISO particle simulations. This code is used to calculate fuel temperature, kernel swelling, buffer densification, thermal and irradiation creep, fracture, and fission gas production and release. One of the recent additions to the BISON code is the ability to model the process of layer debonding. This paper will focus on the simulation results of the improved BISON debonding model that will utilize updated strengths measured from irradiated AGR TRISO fuel particles. The new experimental strength data from micromechanical tests of irradiated TRISO fuel samples were exercised in the BISON simulations and compared to baseline strength data to assess their applicability in the models. This also includes updated buffer-IPyC bond strengths to simulate layer delamination. Based on current experimental observations it is noted that the buffer-IPyC separation occurs not exactly at the junction of these two layers, but more on the side of the buffer layer. That observation is also implemented in the TRISO interface debonding model. This improved modeling approach using experimental strength data to characterize buffer-IPyC debonding and its potential subsequent cracking will be presented in the paper along with comparisons to available experimental observations.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2022

Instantiation of HCML Demonstrating Bayesian Predictive Modeling for Attentional Control

The research team developed models of Attentional Control (AC) that are unique to existing modeling approaches in the literature. The goal was to enable the research team to (1) make predictions about AC and human performance in real-world scenarios and (2) to make predictions about individual characteristics based on human data. First, the team developed a proof-of-concept approach for representing an experimental design and human subjects data in a Bayesian model, then demonstrated an ability to draw inferences about conditions of interest relevant to real-world scenarios. Ultimately, this effort was successful, and we were able to make reasonable (meaning supported by behavioral data) inferences about conditions of interest to develop a risk model for AC (where risk is defined as a mismatch between AC and attentional demand). The team additionally defined a path forward for a human-constrained machine learning (HCML) approach to make predictions about an individual's state based on performance data. The effort represents a successful first step in both modeling efforts and serves as a basis for future work activities. Numerous opportunities for future work have been defined.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Ecological niche models display nonlinear relationships with abundance and demographic performance across the latitudinal distribution of Astragalus utahensis (Fabaceae)

Abstract The potential for ecological niche models (ENMs) to accurately predict species' abundance and demographic performance throughout their geographic distributions remains a topic of substantial debate in ecology and biogeography. Few studies simultaneously examine the relationship between ENM predictions of environmental suitability and both a species' abundance and its demographic performance, particularly across its entire geographic distribution. Yet, studies of this type are essential for understanding the extent to which ENMs are a viable tool for identifying areas that may promote high abundance or performance of a species or how species might respond to future climate conditions. In this study, we used an ensemble ecological niche model to predict climatic suitability for the perennial forb Astragalus utahensis across its geographic distribution. We then examined relationships between projected climatic suitability and field‐based measures of abundance, demographic performance, and forecasted stochastic population growth (λ s ). Predicted climatic suitability showed a J‐shaped relationship with A. utahensis abundance, where low‐abundance populations were associated with low‐to‐intermediate suitability scores and abundance increased sharply in areas of high predicted climatic suitability. A similar relationship existed between climatic suitability and λ s from the center to the northern edge of the latitudinal distribution. Patterns such as these, where density or demographic performance only increases appreciably beyond some threshold of climatic suitability, support the contention that ENM‐predicted climatic suitability does not necessarily represent a reliable predictor of abundance or performance across large geographic regions.

Baer, Kathryn C.↗
DOE OSTI2023

Stream Temperature Prediction in a Shifting Environment: Explaining the Influence of Deep Learning Architecture

Stream temperature is a fundamental control on ecosystem health. Recent efforts incorporating process guidance into deep learning models for predicting stream temperature have been shown to outperform existing statistical and physical models. This performance is in part because deep learning architectures can actively learn spatiotemporal relationships that govern how water and energy propagate through a river network. However, exploration of how spatiotemporal awareness and process guidance influence a model's generalizability under shifting environmental conditions such as climate change is limited. Here, we use Explainable Artificial Intelligence (XAI) to interrogate how differing deep learning architectures affect a model's learned spatial and temporal dependencies, and how those learned dependencies affect a model's ability to maintain high accuracy when applied to unseen environmental conditions. Using the Delaware River Basin in the northeastern United States as a test case, we compare two spatiotemporally aware process–guided deep learning models for predicting stream temperature (a recurrent graph convolution network—RGCN, and a temporal convolution graph model—Graph WaveNet). Both models achieve equally high predictive performance when testing data are well represented in the training data (test root mean squared errors of 1.64°C and 1.65°C); however, Graph WaveNet significantly outperforms RGCN in 4 out of 5 experiments where test partitions represent different types of unseen environmental conditions. XAI results show that the architecture of Graph WaveNet leads to learned spatial relationships with greater fidelity to physical processes, and that this fidelity improves the generalizability of the model when applied to shifting and/or unseen environmental conditions.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Large-Eddy Simulations of Idealized Atmospheric Boundary Layers Using Nalu-Wind

Accurate prediction of wind-plant performance relies, in part, on properly characterizing the turbulent atmospheric boundary layer (ABL) flow in which wind turbines operate. Large-eddy simulation (LES) is a powerful tool for simulating ABLs because it resolves the largest, most energetic scales of three-dimensional turbulent motions. Yet LES predictions are well known to depend on modeling choices such as grid resolution, numerical discretization schemes, and closures for unresolved scales of turbulence. Here, we evaluate how these choices influence predictions of ABL winds using Nalu-Wind, a wind-specific fork of the open-source, generalized, unstructured, massively parallel flow solver NaluCFD/Nalu.

17 WIND ENERGY↗
DOE OSTI2025

Enhanced Feedstock Characterization and Modeling to Facilitate Optimal Preprocessing and Deconstruction of Corn Stover (Final Report)

This project addresses the challenge of processing corn stover by fractionating this biomass feedstock to both streamline processing and generate new potential co-products. Additionally, the project developed new field-deployable analytical tools that can be coupled with empirical models that were used to predict feedstock properties and processing performance. The overall scope of this project was: (1) identify conditions for optimal corn stover fractionation using a two- stage physical fractionation, (2) assess how physical fractionation impacts properties, partitioning of biomass, and response to processing, (3) further adapt, develop, and validate several advanced characterization tools for assessing biomass properties that can be linked to processing behavior, and (4) develop and validate predictive models based on measurements that can be performed “in the field” or “at the biorefinery gate” to predict feedstock processing behavior (preprocessing and deconstruction). The first objective employed pre-separation processing (size reduction) which was next subjected to enhanced separations to yield fractions enriched or depleted in select compositional components or properties. For the second objective, fractions were screened for their response to post-separation processing (pretreatment and enzymatic hydrolysis). Detailed characterization profiles were developed and dynamic image analysis to assess distribution of particle size and morphology. For the final objective, we utilized these tools to develop empirical models to assess the relative abundance of tissue type in order to assess fractionation efficacy and to predict fraction performance during pretreatment and enzymatic hydrolysis.

09 BIOMASS FUELS↗
DOE OSTI2020

Toward particle-resolved accuracy in Euler–Lagrange simulations of multiphase flow using machine learning and pairwise interaction extended point-particle (PIEP) approximation

This study presents two different machine learning approaches for the modeling of hydrodynamic force on particles in a particle-laden multiphase flow. Results from particle-resolved direct numerical simulations (PR-DNS) of flow over a random array of stationary particles for eight combinations of particle Reynolds number (Re) and volume fraction (φ) are used in the development of the models. The first approach follows a two-step process. In the first flow prediction step, the perturbation flow due to a particle is obtained as an axisymmetric superposable wake using linear regression. In the second force prediction step, the force on a particle is evaluated in terms of the perturbation flow induced by all its neighbors using the generalized Faxén form of the force expression. In the second approach, the force data on all the particles from the PR-DNS simulations are used to develop an artificial neural network (ANN) model for direct prediction of force on a particle. Due to the unavoidable limitation on the number of fully resolved particles in the PR-DNS simulations, direct force prediction with the ANN model tends to over-fit the data and performs poorly in the prediction of test data. In contrast, due to the millions of grid points used in the PR-DNS simulations, accurate flow prediction is possible, which then allows accurate prediction of particle force. This hybridization of multiphase physics and machine learning is particularly important, since it blends the strength of each, and the resulting pairwise interaction extended point-particle model cannot be developed by either physics or machine learning alone.

42 ENGINEERING↗
DOE OSTI2026

Dominant Controls on Preferential Flow and Their Implications for Future Soil Water Fluxes

Abstract Soil water flow, particularly preferential flow (PF), is a critical control on hydrological and biogeochemical processes, including groundwater recharge, contaminant transport, and carbon cycling. However, it remains challenging to predict PF occurrence across large environmental gradients. Here, we developed a deep learning (DL) model to estimate event‐scale soil water flow velocity and the probability of PF occurrence using high‐frequency soil moisture and precipitation data from 33 sites across the National Ecological Observatory Network. The model demonstrated high skill in predicting the binary occurrence of PF (91% F1‐score; 85% accuracy) but the performance was limited in predicting soil water velocity ( R 2 = 0.31). We found that precipitation characteristics (duration, volume, and intensity) were the most important predictors for soil water velocity. Among the non‐precipitation event variables, sand content showed relatively high predictive skill, though differences among non‐event climate variables were generally modest. Lower sand content was associated with increased predicted soil water velocity, a finding that highlights the role of soil structure in producing more non‐uniform flow, which contrasts with traditional uniform flow models. Projecting a reduced DL model under both moderate and high‐emissions future climate scenarios (2060–2099 Representative Concentration Pathways 4.5 and 8.5), we found ∼7.3% increase under RCP4.5 and ∼15% under RCP8.5 of soil water velocities compared to the historical simulation, while modeled likelihood of PF changed little. These findings suggest climate change is not making PF more frequent, but it is making existing PF pathways more efficient with important consequences for associated nutrient and contaminant transport under climate change. Plain Language Summary Water movement in soil is critical for water quality. While often modeled as a uniform flow process, in reality water moves rapidly through cracks and burrows in what is called “preferential flow” (PF), which limits natural filtration and can transport pollutants. We developed a deep learning model, trained on data from 33 U.S. sites, to predict when and how fast this PF occurs based on precipitation, soil, and climate data. The model showed that precipitation characteristics (duration, intensity, volume) were the most important predictors of PF. Lower soil sand content/higher clay content was associated with faster water flow, likely due to clay soils forming aggregates and cracks that water moves through rather than infiltrating uniformly. Further analyses based on climate projections suggest that the speed at which PF occurs will become more rapid under future climate scenarios compared to historical simulation. This highlights the need to represent PF in soil water models when assessing future water quality. Key Points The effect of precipitation peak intensity on soil water velocities declined with increasing precipitation intensity Antecedent soil moisture failed to predict preferential flow (PF), contrasting the high predictive power of sand content Climate predictions suggest that soil water velocities through PF paths will increase ∼15% by 2099

Li, Bonan↗
DOE OSTI2023

Editorial: Sorption processes in nuclear waste management: data knowledge management and new methodologies for data acquisition/prediction

A fundamental approach to Nuclear Waste Repository research involves the collection of experimental data in a laboratory setting, development of empirical and/or mechanistic numerical models representing those observations, and application of these models (or Reduced Order Models) into reactive transport and performance assessment models as predictive tools for informing society of impacts and risks associated with nuclear waste repository scenarios (Stevens et al., 2020). Therefore, the assimilation and interpretation of experimental data must take advantage of both new data and the rich historical data available in the literature and apply novel modeling approaches to improve predictive tools, particularly from the standpoint of uncertainty quantification, for nuclear waste repository performance assessment (Zavarin et al., 2022).

sorption↗
DOE OSTI2024

Feedstock variability impacts the bioconversion of sugar and lignin streams derived from corn stover by Clostridium tyrobutyricum and engineered Pseudomonas putida

Abstract Feedstock variability represents a challenge in lignocellulosic biorefineries, as it can influence both lignocellulose deconstruction and microbial conversion processes for biofuels and biochemicals production. The impact of feedstock variability on microbial performance remains underexplored, and predictive tools for microbial behaviour are needed to mitigate risks in biorefinery scale‐up. Here, twelve batches of corn stover were deconstructed via deacetylation, mechanical refining, and enzymatic hydrolysis to generate lignin‐rich and sugar streams. These batches and their derived streams were characterised to identify their chemical components, and the streams were used as substrates for producing muconate and butyrate by engineered Pseudomonas putida and wildtype Clostridium tyrobutyricum , respectively. Bacterial performance (growth, product titers, yields, and productivities) differed among the batches, but no strong correlations were identified between feedstock composition and performance. To provide metabolic insights into the origin of these differences, we evaluated the effect of twenty‐three isolated chemical components on these microbes, including three components in relevant bioprocess settings in bioreactors, and we found that growth‐inhibitory concentrations were outside the ranges observed in the streams. Overall, this study generates a foundational dataset on P. putida and C. tyrobutyricum performance to enable future predictive models and underscores their resilience in effectively converting fluctuating lignocellulose‐derived streams into bioproducts.

09 BIOMASS FUELS↗
DOE OSTI2024

Machine learning assisted prediction of tungsten heavy alloy plasma facing component performance for fusion energy applications

Tungsten and tungsten heavy alloys (WHAs), known for their remarkably high hardness, durability, and corrosion resistance, play a critical role in the thriving development of nuclear fusion reactors in recent years. However, the exploration in tungsten alloys for the nuclear-related applications has been limited by the difficulty of manufacturing and the complexity of experiments to reproduce the environment of nuclear reaction. Therefore, this project aims to utilize nanoscale simulation methods such as density functional theory (DFT) and molecular dynamics (MD) with the help of machine learning techniques to not only understand the mechanisms of tungsten alloys but also allow us to computationally predict their mechanical behaviors under extreme environments. One critical problem of the application of WHAs in nuclear reactors is the surface melting. In the current design of the SPARC reactor, the WHA, W97Ni2.1Fe0.9 or W97NiFe, is chosen to be the first wall components to confine the plasma where the particles are fiercely moving and colliding into each other to create nuclear fusion reaction. This process will generate extremely high heat flux onto these WHA tiles, leaving high surface temperature that could possibly melt the surface of the WHA tiles, As illustrated in Fig. 1(a). a laser experiment previously done illustrates that a rough surface damage would be made after the surface melting where the matrix area mainly composed of nickel and iron as shown in Fig. 1(b), will first melt and then leave vacancies between these tungsten grains. Unfortunately, these kinds of roughness on the first-wall components could be deadly to the plasma inside a Tokmak reactor because the heat that is supposed to dissipate at a designed ratio through the tiles may in turn be excessively absorbed and accumulated on any uneven area of the surface, which will eventually make the whole nuclear reaction fail. In this project, we will introduce a machine learning potential, Allegro, based on DFT calculation and then build a MD model for W-Ni-Fe alloys.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Constitutive Model Development for Aging Polymer Encapsulants (ASC P&EM FY2021 L2 Milestone 7836)

This SAND report fulfills the completion requirements for the ASC Physics and Engineering Modeling Level 2 Milestone 7836 during Fiscal Year 2021. The Sandia Simplified potential energy clock (SPEC) non-linear viscoelastic constitutive model was developed to predict a whole host of polymer glass physical behaviors in order to provide a tool to assess the effects of stress on these materials over their lifecycle. Polymer glasses are used extensively in applications such as electronics packaging, where encapsulants and adhesives can be critical to device performance. In this work, the focus is on assessing the performance of the model in predicting material evolution associated with long-term physical aging, an area that the model has not been fully vetted in. These predictions are key to utilizing models to help demonstrate electronics packaging component reliability over decades long service lives, a task that is very costly and time consuming to execute experimentally. The initiating hypothesis for the work was that a model calibration process can be defined that enables confidence in physical aging predictions under ND relevant environments and timescales without sacrificing other predictive capabilities. To test the hypothesis, an extensive suite of calibration and aging data was assembled from a combination of prior work and collaborating projects (Aging and Lifetimes as well as the DoD Joint Munitions Program) for two mission relevant epoxy encapsulants, 828DGEBA/DEA and 828DGEBA/T403. Multiple model calibration processes were developed and evaluated against the entire set of data for each material. A qualitative assessment of each calibration's ability to predict the wide range of aging responses was key to ranking the calibrations against each other. During this evaluation, predictions that were identified as non-physical, i.e., demonstrated something that was qualitatively different than known material behavior, were heavily weighted against the calibration performance. Thus, unphysical predictions for one aspect of aging response could generate a lower overall rating for a calibration process even if that process generated better quantitative predictions for another aspect of aging response. This insurance that all predictions are qualitatively correct is important to the overall aim of utilizing the model to predict residual stress evolution, which will depend on the interplay amongst the different material aging responses. The DSC-focused calibration procedure generated the best all-around aging predictions for both materials, demonstrating material models that can qualitatively predict the whole host of different physical aging responses that have been measured. This step forward in predictive capability comes from an unanticipated source, utilization of calorimetry measurements to specify model parameters. The DSC-focused calibration technique performed better than compression-focused techniques that more heavily weigh measurements more closely related to the structural responses to be predicted. Indeed, the DSC-focused calibration procedure was only possible due to recent incorporation of the enthalpy and heat capacity features into SPEC that was newly verified during this L2 milestone. Fundamentally similar aspects of the two material model calibrations as well as parametric studies to assess sensitives of the aging predictions are discussed within the report. A perspective on the next steps to the overall goal of residual stress evolution predictions under stockpile conditions closes the report.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Performance Comparison of Machine Learning Models for Ultrasonic Nondestructive Evaluation of Alkali-Silica Reaction in Concrete

Alkali-silica reaction (ASR) causes concrete degradation, leading to cracking, rebar corrosion, and reduced structural integrity, which raises safety concerns. Ultrasonic nondestructive evaluation (NDE) effectively assesses concrete properties and monitors ASR progression. However, its deployment and analysis require specialized expertise and subjective interpretation. As computational power increases, artificial intelligence (AI) and machine learning (ML) algorithms are increasingly being used to automate NDE data analysis across various industries for AI-assisted automation. Regulatory agencies are adapting to this technological shift, prompting a need to evaluate current ML technologies’ capabilities and limitations in assessing concrete material properties and damage. This report presents a comparative analysis of four ML regression models for predicting concrete material damage induced by ASR expansion using long-term ultrasonic data monitoring. The models investigated include linear regression (LR), support vector regression (SVR), shallow neural networks (NN), and deep neural networks (DNN). LR, SVR, and shallow NN models use features extracted from ultrasonic signals, whereas the DNN model processes time-domain ultrasonic signals and frequency spectra directly. The study systematically compared the models’ performance from various perspectives, including model input, prediction performance, and generalization ability. The findings indicate significant variability in model performance, with some ML algorithms achieving very high or very low prediction accuracy depending on the preprocessing and feature engineering (extraction and selection) applied. Key insights include the observation that shallow ML models (LR, SVR, and shallow NNs) require meticulous preprocessing and feature extraction to achieve high accuracy. In contrast, the DNN model, although it bypasses the need for feature engineering, necessitates extensive preprocessing to mitigate noise and computational demands. The SVR model emerged as the top performer among the shallow models, and the DNN model exhibited superior performance on specific datasets but struggled with generalization across specimens from different batches. Additionally, the SVR model is sensitive to temperature variations, whereas the DNN model is robust in this regard. Using recurrent neural networks is recommended for future ASR expansion prediction studies. Recurrent neural networks’ inherent ability to capture temporal dependencies and long-term patterns makes them well suited for analyzing sequential ultrasonic monitoring data. Overall, the results and conclusions of this study could provide insights into the capabilities and effectiveness of ML when applied to ultrasonic NDE data and help identify best practices for using ML for ultrasonic NDE of concrete material properties.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Materials Characterization, Prediction and Control Project: Summary Report on Data Analytics Framework

This report summarizes the activities performed under the data analytics Vertex in the Materials Characterization, Prediction and Control Project funded under laboratory directed research and development at Pacific Northwest National Laboratory. The data analytics Vertex developed models for associating global or local process parameters, microstructural features, and performance properties of friction-stir-processed 316L stainless steel plates. Statistical, machine learning, and deep learning models, as well as generative artificial intelligence approaches, were used to develop the associations between the process-structure-property data streams. These associations formed the basis for predicting global properties of parts manufactured under different process envelopes, providing a basis for predicting performance using data driven as well as physics-informed and physics-constrained approaches. Additionally, the associations were used to predict local process parameters and microstructural features of the product, predictive relationships that have the potential to form the basis of a control framework that could eventually modulate a friction-stir process to maintain product quality.

316L stainless steel↗
DOE OSTI2023

Machine Learning for Daily Forecasts of Arctic Sea Ice Motion: An Attribution Assessment of Model Predictive Skill

Physics-based simulations of Arctic sea ice are highly complex, involving transport between different phases, length scales, and time scales. Resultantly, numerical simulations of sea ice dynamics have a high computational cost and model uncertainty. We employ data-driven machine learning (ML) to make predictions of sea ice motion. The ML models are built to predict present-day sea ice velocity given present-day wind velocity and previous-day sea ice concentration and velocity. Models are trained using reanalysis winds and satellite-derived sea ice properties. We compare the predictions of three different models: persistence (PS), linear regression (LR), and a convolutional neural network (CNN). We quantify the spatiotemporal variability of the correlation between observations and the statistical model predictions. Additionally, we analyze model performance in comparison to variability in properties related to ice motion (wind velocity, ice velocity, ice concentration, distance from coast, bathymetric depth) to understand the processes related to decreases in model performance. Results indicate that a CNN makes skillful predictions of daily sea ice velocity with a correlation up to 0.81 between predicted and observed sea ice velocity, while the LR and PS implementations exhibit correlations of 0.78 and 0.69, respectively. The correlation varies spatially and seasonally: lower values occur in shallow coastal regions and during times of minimum sea ice extent. LR parameter analysis indicates that wind velocity plays the largest role in predicting sea ice velocity on 1-day time scales, particularly in the central Arctic. Regions where wind velocity has the largest LR parameter are regions where the CNN has higher predictive skill than the LR.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Comparative Analysis of TCR and TCR-pMHC Complex Structure Prediction Tools

The rapid development of computational approaches for predicting the structures of T cell receptors (TCRs) and TCR-peptide-major histocompatibility (TCR-pMHC) complexes, accelerated by AI breakthroughs such as AlphaFold, has made it feasible to calculate these structures with increasing accuracy. Although these tools show great potential, their relative accuracy and limitations remain unclear due to the lack of standardized benchmarks. Here, we systematically evaluate seven tools for predicting isolated TCR structures together with six tools for predicting TCR-pMHC complex structures. The methods include homology-based approaches, general prediction tools using AlphaFold, TCR-specific tools derived from AlphaFold2, and the newly developed tFold-TCR model. The evaluation uses a post-training data set comprising 40 αβ TCRs and 27 TCR-pMHC complexes (21 Class I and 6 Class II). Model accuracy is assessed at global, local, and interface levels using a variety of metrics. We find that each tool offers distinct advantages in various aspects of its predictions. AlphaFold2, AlphaFold3, and tFold-TCR excel in overall accuracy of TCR structure prediction, and TCRmodel2 and AlphaFold2 perform well in overall accuracy of TCR-pMHC structure prediction. However, TCR-specific tools derived from AlphaFold2 show lower accuracy in the framework region than both homology-based methods and general-purpose tools such as AlphaFold, and challenges remain for all in modeling CDR3 loops, docking orientations, TCR-peptide interfaces, and Class II MHC-peptide interfaces. Furthermore, these findings will guide researchers in selecting appropriate tools, emphasize the importance of using multiple evaluation metrics to assess model performance, and offer suggestions for improving TCR and TCR-pMHC structure prediction tools.

Chemical structure↗