SEARCH · Engineering Papers

Results for “prediction methods”

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 523 records · Page 29

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Simulation Tools for Characterizing Stress Distribution in Laser Welded Dissimilar Joints

This project focuses on developing a thermo-metallurgical-mechanical modeling method to accurately predict the microstructural evolution and residual stress in laser welding between dissimilar metals, such as HSLA steel and high carbon equivalent (CE) gear steel. The method leverages a comprehensive material database to model the temperature and rate dependent phase transformations, along with their associated effects on material properties, such as thermal expansion and flow stress, throughout the welding process. A key innovation is the incorporation of phase transformation and phase-specific properties, which enhances the accuracy of residual stress predictions. The mixture material in the fusion zone due to the dissimilar metals will also be addressed in the numerical model. This is especially critical in scenarios involving phase transformations in the fusion zone and heat-affected zone (HAZ), where the phase changes can induce substantial residual stress variations. The material database has been generated using JMatPro. The modeling approach is implemented through a custom User Material (UMAT) subroutine, executed with the commercial finite element software Abaqus.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Simulation Tools for Characterizing Stress Distribution in Laser Welded Dissimilar Joints

36 MATERIALS SCIENCE↗

DOE OSTI2024

Errant Beam Prognostics with Machine Leaning at SNS Accelerator

Particle Accelerators are complex machine with many pieces of equipment running in synchronization to deliver required beam. However, faults in particle accelerators reduce the availability of the beam for experiments affecting the overall science output. To avoid these faults, we apply anomaly detection techniques to predict any unusual behavior and perform preemptive actions to improve the total availability. Many researchers have adopted semi-supervised Machine Learning (ML) methods such as auto-encoders and variational auto-encoders for such tasks. However, supervised ML techniques designed for similarity learning such as Siamese Neural Network (SNN) can outperform semi-supervised or unsupervised methods for anomaly prediction. One of the challenges associated with application of ML models to particle accelerators is the variability in observed data over time due to system configuration changes. We employ conditional models such as Conditional Siamese Neural Networks (CSNN), and Conditional-VAE (CVAE) to learn the variability in the data by using beam configuration parameters as conditional input. We apply these models for errant beam prediction at Spallation Neutron Source accelerator under different system configurations and compare their performance. We demonstrate that CSNN outperforms CVAE in our application. This talk will present the data source, collection, analysis, data-preparation, model development, hyper-parameter studies and the results.

Rajput, Kishansingh↗

DOE OSTI2025

Decayheatml

This code is designed to predict and analyze the decay heat generated in molten salt reactors (MSRs) using a hybrid approach that combines machine learning and segmented polynomial fitting. The accurate prediction of decay heat is essential for reactor safety and the optimization of spent fuel storage. The code operates through several key components: 1) Data Architecture: It incorporates a modular data architecture that handles various MSR-specific operational parameters such as power density, humidity content, and air ingress. These parameters are sampled using Sobol sequences to ensure comprehensive coverage of operational uncertainties. 2) Machine Learning Framework: The code employs a diverse set of machine learning models, including polynomial regression, decision trees, random forests, gradient boosting, support vector regression, k-nearest neighbors, multi-layer perceptrons, and symbolic regression. These models are trained to predict decay heat over a wide temporal range, from immediate shutdown up to 10,000 years. 3) Region-Optimized Training: The temporal domain is divided into multiple regions, each modeled separately to capture distinct decay heat characteristics across different time scales. This approach significantly improves the accuracy and interpretability of predictions. 4) Segmented Polynomial Interpretation (SPI): The SPI method translates machine learning predictions into piecewise polynomial equations. These equations are physically interpretable and can be directly integrated into existing engineering workflows and safety analyses. 5) Front-End Interfaces: The code includes both a Jupyter notebook interface for research development and a Streamlit web application for operational deployment. These interfaces allow users to interactively explore decay heat predictions, adjust operational parameters, and visualize results in real-time. 6) Applications: The framework supports various applications, including safety system validation and spent fuel container optimization. It enables real-time evaluation of worst-case decay heat scenarios, informing the design of passive safety systems and optimizing container designs for long-term storage. Overall, this code provides a robust, accurate, and user-friendly tool for predicting decay heat in MSRs, enhancing reactor safety, and optimizing spent fuel management.

Retamales, Mauricio Eduardo Tano [Idaho National L↗

DOE OSTI2024

Constraining Galaxy-Halo connection using machine learning

We investigate the potential of machine learning (ML) methods to model small-scale galaxy clustering for constraining Halo Occupation Distribution (HOD) parameters. Our analysis reveals that while many ML algorithms report good statistical fits, they often yield likelihood contours that are significantly biased in both mean values and variances relative to the true model parameters. This highlights the importance of careful data processing and algorithm selection in ML applications for galaxy clustering, as even seemingly robust methods can lead to biased results if not applied correctly. ML tools offer a promising approach to exploring the HOD parameter space with significantly reduced computational costs compared to traditional brute-force methods if their robustness is established. Using our ANN-based pipeline, we successfully recreate some standard results from recent literature. Properly restricting the HOD parameter space, transforming the training data, and carefully selecting ML algorithms are essential for achieving unbiased and robust predictions. Among the methods tested, artificial neural networks (ANNs) outperform random forests (RF) and ridge regression in predicting clustering statistics, when the HOD prior space is appropriately restricted. We demonstrate these findings using the projected two-point correlation function (w p (r p )), angular multipoles of the correlation function (ξ ℓ (r)), and the void probability function (VPF) of Luminous Red Galaxies from Dark Energy Spectroscopic Instrument mocks. Our results show that while combining w p (r p ) and VPF improves parameter constraints, adding the multipoles ξ 0 , ξ 2 , and ξ 4 to w p (r p ) does not significantly improve the constraints.

cosmology↗

DOE OSTI2025

Uncertainty quantification for nuclear forensics with population analyses

Although neural networks offer cutting-edge predictive power, their deployment in high-consequence nuclear forensic applications is limited, partly because of their black-box nature. Incorporating robust uncertainty quantification methods into the predictive frameworks of neural networks is progress towards their future deployment in such scenarios. This work integrates uncertainty quantification into neural networks for nuclear reactor core-average burnup estimation from simulated environmental samples. We test two regimes (homogeneous and heterogeneous events) on DeepSets and Set Transformer architectures, we find both quantify predictive uncertainty effectively, but Set Transformer excels in partitioning latent events, offering superior predictive power and more informative uncertainty estimates.

Hatton, Conner [ORNL] (ORCID:0009000804970959)↗

DOE OSTI2021

Selecting durable building envelope systems with machine learning assisted hygrothermal simulations database

Hygrothermal simulations provide insight into the energy performance and moisture durability of building envelope components under dynamic conditions. The inputs required for hygrothermal simulations are extensive, and carrying out simulations and analyses requires expert knowledge. An expert system, the Building Science Advisor (BSA), has been developed to predict the performance and select the energy-efficient and durable building envelope systems for different climates. The BSA consists of decision rules based on expert opinions and thousands of parametric simulation results for selected wall systems. The number of potential wall systems results in millions, too many to simulate all of them. We present how machine learning can help predict durability data, such as mold growth, while minimizing the number of simulations needed to run. The simulation results are used for training and validation of machine learning tools for predicting wall durability. We tested Artificial Neural Network (ANN) and Gradient Boosted Decision Trees (GBDT) for their applicability and model accuracy. Models developed with both methods showed adequate prediction performance (root mean square error of 0.195 and 0.209, respectively). Finally, we introduce how the information supports guidance for envelope design via an easy-to-use web-based tool that does not require the end-user to run hygrothermal simulations.

Salonvaara, Mikael↗

DOE OSTI2024

Experimental validation of a collision-radiation dataset for molecular hydrogen in plasmas

Quantitative spectroscopy of molecular hydrogen has generated substantial demand, leading to the accumulation of diverse elementary process data encompassing radiative transitions, electron-impact transitions, predissociations, and quenching. However, their rates currently available are still sparse, and there are inconsistencies among those proposed by different authors. In this study, we demonstrate an experimental validation of such a molecular dataset by composing a collisional-radiative model (CRM) for molecular hydrogen and comparing experimentally obtained vibronic populations across multiple levels. From the population kinetics of molecular hydrogen, the importance of each elementary process in various parameter space is studied. In low-density plasmas (electron density ne≲1017 m−3) the excitation rates from the ground states and radiative decay rates, both of which have been reported previously, determine the excited state population. The inconsistency in the excitation rates affects the population distribution the most significantly in this parameter space. However, in higher density plasmas (ne≳1018 m−3), the excitation rates from excited states become important, which have never been reported in the literature, and may need to be approximated in some way. In order to validate these molecular datasets and approximated rates, we carried out experimental observations for two different hydrogen plasmas; a low-density radio frequency heated plasma (ne≈1016 m−3) and the Large Helical Device (LHD) divertor plasma (ne≳1018 m−3). The visible emission lines from EF1Σg+, HH¯1Σg+, D1Πu±, GK1Σg+, I1Πg±, J1Δg±, h3Σg+, e3Σu+, d3Πu±,g3Σg+, i3Πg±, and j3Δg± states were observed simultaneously and their population distributions were obtained from their intensities. We compared the observed population distributions with the CRM prediction, in particular the CRM with the rates compiled by Janev et al., Miles et al., and those calculated with the molecular convergent close-coupling (MCCC) method. The MCCC prediction gives the best agreement with the experiment, particularly for the emission from the low-density plasma. However, the population distribution in the LHD divertor shows a worse agreement with the CRM than those from low-density plasma, indicating the necessity of the precise excitation rates from excited states. We also found that the rates for the electron attachment is inconsistent with experimental results. This requires further investigation.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2020

Real-Time Movement-Based Traffic Volume Prediction at Signalized Intersections

The traffic volume of each movement at signalized intersections can provide valuable information on real-time traffic conditions that enable traffic control systems to dynamically respond to the fluctuated traffic demands. Real-time movement-based traffic volume prediction is challenging due to various nonlinear spatial relationships at different locations/approaches and the complicated underlying temporal dependencies. In this study, a novel deep intersection spatial-temporal network (DISTN) is developed for real-time movement-based traffic volume prediction at signalized intersections, which considers both spatial and temporal features by the convolutional neural network (CNN) and long short-term memory (LSTM), respectively. In addition, the within-day, daily, and weekly periodic trends of traffic volume are also considered in the proposed model. This is the first time that a deep-learning method has been applied for movement-based traffic volume prediction at signalized intersections. In the numerical experiment, the proposed model is evaluated using real-world data and simulation data to demonstrate its effectiveness. The impacts of various structures of traffic networks on the proposed model are also discussed. Results show that the proposed model outperforms some of the state-of-the-art volume prediction methods currently in the literature.

99 GENERAL AND MISCELLANEOUS↗

DOE OSTI2025

Accurate and Efficient Prediction of p K w in Aqueous Electrolytes Using Local Electrostatic Potentials

The K w =10 -pK w in aqueous electrolytes exhibit significant variations as a function of ion concentration. This behavior is not well-described by the Pitzer model, particularly at high concentrations. Here, we provide a molecular interpretation of the concentration dependent trends in p K w and develop a new method that accurately predicts the dissociation constant based upon the local electrostatic potential of the water oxygen that is determined by its nearest neighbors. The method is computationally efficient, relying only upon sampling of the local configurations (via molecular dynamics) and a small set of experimentally measured p K w . It is extensible to a range of salt compositions and the molecular understanding of the local potentials provides new routes toward the development of electrolytes with tailored physicochemical properties like p K w .

anions↗

DOE OSTI2025

Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models

Abstract We evaluate the ability of machine learning to predict whether a hypothetical crystal structure can be synthesized and explain those predictions to scientists. Fine‐tuned large language models (LLMs) trained on a human‐readable text description of the target crystal structure perform comparably to previous bespoke convolutional graph neural network methods, but better prediction quality can be achieved by training a positive‐unlabeled learning model on a text‐embedding representation of the structure. An LLM‐based workflow can then be used to generate human‐readable explanations for the types of factors governing synthesizability, extract the underlying physical rules, and assess the veracity of those rules. These explanations can guide chemists in modifying or optimizing non‐synthesizable hypothetical structures to make them more feasible for materials design.

Kim, Seongmin [Department of Chemical and Biologic↗

DOE OSTI2025

Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models

Kim, Seongmin [Department of Chemical and Biologic↗

DOE OSTI2024

A methodology for selection of solid desiccants in energy recovery ventilators

Controlling indoor humidity levels is essential for maintaining acceptable indoor air quality in buildings. The use of energy recovery ventilators (ERVs) is an energy-efficient way to regulate indoor air humidity. Fixed-bed regenerators and rotary wheels are widely used ERVs because of their high sensible and latent effectiveness. These ERVs are made of desiccant-coated substrates, which enable them to transfer moisture between the supply and exhaust air streams. However, the moisture transfer ability of ERVs depends on the physiochemical and sorption properties of desiccants. Extensive, full-scale experiments are required to determine the best desiccant material for these systems. This paper presents a simplified method of selecting suitable desiccant materials for ERVs. The methodology involves important characterization methods, literature correlations for performance prediction, and cost-effective testing methods prior to full-scale testing, and full-scale test methods are discussed in detail. Furthermore, the performance of a few newly derived materials is evaluated and compared with that of conventional desiccants such as silica gel and molecular sieves. The highest latent effectiveness was obtained for composite of super absorbent polymer (SAP) with potassium formate (SAP-HCO2K-50 %), all-polymer porous solid desiccant (APPSD) and metal organic framework (MOF)–MIL–101 (Cr), followed by activated carbon fibre felt (ACFF) Silica sol-LiCl30, SAP, silica gel, MOF–303, and molecular sieve. Researchers and manufacturers would benefit from the proposed methodology and presented data in developing new desiccant materials for ERV applications.

Energy recovery↗

DOE OSTI2022

Recent advances and applications of deep learning methods in materials science

Deep learning (DL) is one of the fastest-growing topics in materials data science, with rapidly emerging applications spanning atomistic, image-based, spectral, and textual data modalities. DL allows analysis of unstructured data and automated identification of features. The recent development of large materials databases has fueled the application of DL methods in atomistic prediction in particular. In contrast, advances in image and spectral data have largely leveraged synthetic data enabled by high-quality forward models as well as by generative unsupervised DL methods. In this article, we present a high-level overview of deep learning methods followed by a detailed discussion of recent developments of deep learning in atomistic simulation, materials imaging, spectral analysis, and natural language processing. For each modality we discuss applications involving both theoretical and experimental data, typical modeling approaches with their strengths and limitations, and relevant publicly available software and datasets. We conclude the review with a discussion of recent cross-cutting work related to uncertainty quantification in this field and a brief perspective on limitations, challenges, and potential growth areas for DL methods in materials science.

36 MATERIALS SCIENCE↗

DOE OSTI2026

Development and implementation of high-throughput proteomic and metabolomics assays by using advanced chromatographic and mass spectrometric systems (CRADA Final Report)

The mission of this CRADA with Agilent was to couple powerful MS platforms (QQQ, IM-QTOFMS) with Agilent’s novel Ultra-High-Performance Liquid Chromatography (UHPLC) fast metabolomic workflows and perform ABF Machine Learning (ML) to generated datasets. Agilent transferred UHPLC methods to PNNL and LBNL and methods were implemented and demonstrated in both labs, achieving total acquisition times of < 10 min. Metabolites analyzed using Agilent’s shared methods included metabolites from central carbon metabolism, common across hosts, and metabolites unique to engineered strains. Standards were acquired in an UHPLC-Drift Tube Ion Mobility Mass Spectrometer (DTIMS) system for the first time within the context of ABF and methods were optimized based on Agilent’s protocols. Samples from ABF hosts Pseudomonas putida, Aspergillus pseudoterreus, Aspergillus niger and Rhodosporidium toruloides were analyzed using the UHPLC-DTIMS platform for a total of 276 runs. A data analysis workflow compatible with the Experimental Data Depot (EDD) and completely shareable was developed for the acquired UHPLC-DTIMS data. Samples were analyzed using a Data Independent Acquisition Approach (DIA), which for most of the standards provided more transitions therefore increasing detection confidence. Using the data acquired by PNNL, LBNL, and Agilent’s specifications from previous ML projects, SNL applied an ensemble ML strategy to pick the best performing model for automated LC-method selection. Finally, with the contribution of the participant labs and Agilent, SNL developed an Automated Method Selection (AMS) software tool to predict the best liquid chromatography method for analysis of any new molecules of interest. Samples with novel pathways and new metabolite targets of interest are generated at a high pace in the ABF. Overall, the project advanced rapid metabolomics by combining liquid chromatography, ion mobility spectrometry, and data-independent mass spectrometry with machine learning. This multidimensional approach uses retention time, collision cross-section, precursor mass, and fragment-ion information to distinguish chemically similar metabolites that can be difficult to resolve using conventional liquid- or gas-chromatography methods. The resulting workflow also provided automated metabolite-identification error estimates, addressing a recognized need for statistical confidence measures in metabolomics.

Petzold, Christopher [Lawrence Berkeley National L↗

DOE OSTI2021

Adaptive Machine Learning for Robust Diagnostics and Control of Time-Varying Particle Accelerator Components and Beams

Machine learning (ML) is growing in popularity for various particle accelerator applications including anomaly detection such as faulty beam position monitor or RF fault identification, for non-invasive diagnostics, and for creating surrogate models. ML methods such as neural networks (NN) are useful because they can learn input-output relationships in large complex systems based on large data sets. Once they are trained, methods such as NNs give instant predictions of complex phenomenon, which makes their use as surrogate models especially appealing for speeding up large parameter space searches which otherwise require computationally expensive simulations. However, quickly time varying systems are challenging for ML-based approaches because the actual system dynamics quickly drifts away from the description provided by any fixed data set, degrading the predictive power of any ML method, and limits their applicability for real time feedback control of quickly time-varying accelerator components and beams. In contrast to ML methods, adaptive model-independent feedback algorithms are by design robust to un-modeled changes and disturbances in dynamic systems, but are usually local in nature and susceptible to local extrema. In this work, we propose that the combination of adaptive feedback and machine learning, adaptive machine learning (AML), is a way to combine the global feature learning power of ML methods such as deep neural networks with the robustness of model-independent control. We present an overview of several ML and adaptive control methods, their strengths and limitations, and an overview of AML approaches.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Effect of Grain Boundary Misorientation on Spall Strength in Ta via Shock-Free Simulations with Relatively Few Atoms

A suite of 37 molecular dynamics simulations is conducted at two system sizes to systematically characterize the role of grain boundary (GB) misorientation on spall strength in pure BCC tantalum (Ta). The systems studied consist of bicrystals with a single [110] symmetric tilt grain boundary. Two loading conditions are compared: (i) homogeneous extension under uniaxial strain simulated in this study and (ii) piston/flyer impact of sample, which induces heterogeneous deformation via shockwave propagation along the length of the sample. The piston/flyer impact is taken from the literature and run on the same set of GB misorientation angles using LAMMPS. The major finding here is that both methods result in similar spall strength predictions, but the homogeneous extension method generally requires two to three orders of magnitude fewer atoms and similar reductions in computational costs. Spall strength results systematically overpredict using this method, by about 10% for the dataset three orders of magnitude smaller than piston/flyer simulations, and 5% for the dataset two orders of magnitude smaller. Lastly, the effect of system size and pre-compression magnitude on spall strength is systematically characterized.

36 MATERIALS SCIENCE↗

DOE OSTI2022

Local correlation-based transition models for high-Reynolds-number wind-turbine airfoils

Abstract. Modern wind-turbine airfoil design requires robust performance predictions for varying thicknesses, shapes, and appropriate Reynolds numbers. The airfoils of current large offshore wind turbines operate with chord-based Reynolds numbers in the range of 3–15 million. Turbulence transition in the airfoil boundary layer is known to play an important role in the aerodynamics of these airfoils near the design operating point. While the lack of prediction of lift stall through Reynolds-averaged Navier–Stokes (RANS) computational fluid dynamics (CFD) is well known, airfoil design using CFD requires the accurate prediction of the glide ratio (L/D) in the linear portion of the lift polar. The prediction of the drag bucket and the glide ratio is greatly affected by the choice of the transition model in RANS CFD of airfoils. We present the performance of two existing local correlation-based transition models – one-equation model (γ− SA) and two-equation model (γ-Reθt‾- SA) coupled with the Spalart–Allmaras (SA) RANS turbulence model – for offshore wind-turbine airfoils operating at a high Reynolds number. We compare the predictions of the two transition models with available experimental and CFD data in the literature in the Reynolds number range of 3–15 million including the AVATAR project measurements of the DU00-W-212 airfoil. Both transition models predict a larger L/D compared to fully turbulent results at all Reynolds numbers. The two models exhibit similar behavior at Reynolds numbers around 3 million. However, at higher Reynolds numbers, the one-equation model fails to predict the natural transition behavior due to early transition onset. The two-equation transition model predicts the aerodynamic coefficients for airfoils of various thickness at higher Reynolds numbers up to 15 million more accurately compared to the one-equation model. As a result, the two-equation model predictions are more comparable to the predictions from eN transition model. However, a limitation of this model is observed at very high Reynolds numbers of around 12–15 million where the predictions are very sensitive to the inflow turbulent intensity. The combination of the two-equation transition model coupled with the Spalart–Allmaras (SA) RANS turbulence model is a good method for performance prediction of modern wind-turbine airfoils using CFD.

17 WIND ENERGY↗