Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “predictive analytics”

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 541 records · Page 30
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Effect of Specimen Thickness on Fracture Toughness and Plane Stress to Plane Strain Transition in Medium-Density Polyethylene

This study investigates the effect of specimen thickness on fracture toughness and the transition from plane stress to plane strain conditions in Medium-Density Polyethylene (MDPE) using Single Edge Notched Bend (SENB) specimens. Three thickness groups (7.5 mm, 9.0 mm, and 12.1 mm) were tested following ASTM D5045 protocol. Conditional stress intensity factors (KQ) increased from 2.2 MPavm to 2.8 MPavm with increasing thickness, demonstrating significant size dependency. Confocal microscopy revealed a 34% reduction in maximum crack tip opening displacement (CTOD) from 0.478 mm to 0.314 mm as thickness increased, with plastic zone lateral extent decreasing by 32%. This quantitative evidence validates Irwin's theoretical prediction of plastic zone size transition from plane stress r_y˜ (1/2p) (K_I/s_y )^2 to plane strain conditions r_y˜ (1/6p) (K_I/s_y )^2.Scanning electron microscopy confirmed progressive suppression of shear lips and evolution from ductile tearing with extensive polymer chain drawing in thin specimens to localized crazing in thick specimens. Despite all specimens satisfying the ASTM D5045 criterion (B = 2.5(KQ/s_y )^2 ) ,residual shear lips persisted even in the thickest specimens, demonstrating that nominal size requirements alone are insufficient for achieving complete plane strain conditions in highly ductile polymers. The findings emphasize the necessity of integrated analytical and morphological assessment for valid fracture toughness characterization, with critical implications for life assessment and integrity prediction in thick-section polymer components such as natural gas distribution pipelines.

Medium-Density Polyethylene↗
DOE OSTI2025

Beam focusing and consequences for Doppler backscattering measurements

The phenomenon of focusing of microwave beams in a plasma near a turning-point caustic is discussed by exploiting the analytical solution to the Gaussian beam-tracing equations in the two-dimensional (2-D) linear-layer problem. The location of maximum beam focusing and the beam width at that location are studied in terms of the beam initial conditions. This focusing must be taken into account to interpret Doppler backscattering (DBS) measurements. We find that the filter function that characterises the scattering intensity contribution along the beam path through the plasma is inversely proportional to the beam width, predicting enhanced scattering from the beam focusing region. We show that the DBS signal enhancement for decreasing incident angles between the beam path and the density gradient is due to beam focusing and not due to forward scattering, as was originally proposed by (Gusakov et al., (Plasma Phys. Contr. Fusion, vol. 56, 2014, p. 0250092014, 2017); Plasma Phys. Rep. vol. 43(6), 2017, pp. 605–613). The analytic beam model is used to predict the measurement of the k y density-fluctuation wavenumber power spectrum via DBS, showing that, in an NSTX-inspired example, the spectral exponent of the turbulent, intermediate-to-high k y density-fluctuation spectrum might be quantitatively measurable via DBS, but not the spectral peak corresponding to the driving scale of the turbulent cascade.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Towards Generalizable and Efficient Circuit Topology Design: A Graph-Transformer-based Surrogate Model with Curriculum Learning

Unlike circuit parameter and sizing optimizations, the automated design of analog circuit topologies poses significant challenges for learning-based approaches. One challenge arises from the combinatorial growth of the topology space with circuit size, which limits the topology optimization efficiency. Moreover, traditional circuit evaluation methods are time-consuming, while the presence of data discontinuity in the topology space makes the accurate prediction of circuit performance exceptionally difficult for unseen topologies. To tackle these challenges, we design a novel Graph-Transformer-based Network (GTN) as the surrogate model for circuit evaluation, offering a substantial acceleration in the speed of circuit topology optimization without sacrificing performance. Our GTN model architecture is designed to embed voltage changes in circuit loops and current flows in connected devices, enabling accurate performance predictions for circuits with unseen topologies. To address the cold start problem when scaling GTN to large-scale circuits, we further introduce a curriculum learning strategy that progressively trains GTN from small-scale to large-scale circuits. This approach enables the model to first learn fundamental physical principles from simpler topologies and gradually adapt to complex configurations, effectively bridging the circuit complexity gap and improving prediction accuracy. Taking the power converter circuit design as an experimental task, our GTN model significantly outperforms an analytical approach and baseline methods directly utilizing graph neural networks. Furthermore, GTN achieves less than 5% relative error and 196× speed-up compared with high-fidelity simulation. Notably, our GTN surrogate model empowers an automatic circuit design framework to discover circuits of comparable quality to those identified through high-fidelity simulation while reducing the time required by up to 98.2%. With curriculum learning, the enhanced GTN achieves a 51% improvement for performance prediction of large-scale circuits compared to the GTN model without this strategy. These advancements establish GTN as a scalable framework for automated analog circuit design across varying circuit complexity levels.

Lu, Haoshu [New Jersey Institute of Technology (NJ↗
DOE OSTI2023

Conotoxin Prediction: New Features to Increase Prediction Accuracy

Conotoxins are toxic, disulfide-bond-rich peptides from cone snail venom that target a wide range of receptors and ion channels with multiple pathophysiological effects. Conotoxins have extraordinary potential for medical therapeutics that include cancer, microbial infections, epilepsy, autoimmune diseases, neurological conditions, and cardiovascular disorders. Despite the potential for these compounds in novel therapeutic treatment development, the process of identifying and characterizing the toxicities of conotoxins is difficult, costly, and time-consuming. This challenge requires a series of diverse, complex, and labor-intensive biological, toxicological, and analytical techniques for effective characterization. While recent attempts, using machine learning based solely on primary amino acid sequences to predict biological toxins (e.g., conotoxins and animal venoms), have improved toxin identification, these methods are limited due to peptide conformational flexibility and the high frequency of cysteines present in toxin sequences. This results in an enumerable set of disulfide-bridged foldamers with different conformations of the same primary amino acid sequence that affect function and toxicity levels. Consequently, a given peptide may be toxic when its cysteine residues form a particular disulfide-bond pattern, while alternative bonding patterns (isoforms) or its reduced form (free cysteines with no disulfide bridges) may have little or no toxicological effects. Similarly, the same disulfide-bond pattern may be possible for other peptide sequences and result in different conformations that all exhibit varying toxicities to the same receptor or to different receptors. We present here new features, when combined with primary sequence features to train machine learning algorithms to predict conotoxins, that significantly increase prediction accuracy.

collisional cross section↗
DOE OSTI2024

Prediction of residual stresses in additively manufactured parts using lumped capacitance and classical lamination theory

Several industries are interested in Laser Powder Bed Fusion (L-PBF) Additively Manufactured (AM) metal parts because their designs can be made arbitrarily complex while retaining bulk-type material properties. However, the residual stresses (RS) and distortions caused by the heat gradients inherent to L-PBF processes are detrimental to the structural integrity of the parts and must be taken into consideration during the part design cycle. Predicting the state of stresses in as-built 3D printed parts is a difficult problem that is typically approached with the use of transient thermomechanical Finite Element Models (FEMs). However, the nonlinearities associated with AM processes are difficult to capture in these FEMs without increasing the computational cost of the simulation, limiting their ability to be incorporated into practical design cycles. This work presents a novel analytical framework that combines lumped capacitance nonlinear heat transfer with time dependent classical lamination theory to efficiently and accurately predict RS in as-built L-PBF parts without the need of FEMs. The simulation was compared to Neutron Diffraction (ND) residual strain measurements taken at Oak Ridge National Laboratories (ORNL) as well as Synchrotron X-ray Diffraction (XRD) strain data published by the National Institute of Standards and Technology (NIST). The simulation predictions and the experimental data showed excellent agreement for the in-plane strain directions, and general agreement for the out of plane strain component, highlighting an area where further development can be implemented.

42 ENGINEERING↗
DOE OSTI2022

Data-driven simultaneous process optimization and adsorbent selection for vacuum pressure swing adsorption

Technologies for post-combustion carbon capture are essential for the reduction of greenhouse gas emissions to the atmosphere. However, they are still associated with high costs and energy consumption. Intensified processes for carbon capture have the potential to overcome these challenges due to their higher efficiency, lower capital cost, and increased operational flexibility. Here, this work investigates simultaneous optimization of process conditions and adsorbent selection for a modular Vacuum Pressure-Swing Adsorption system designed for CO 2 capture. Both surrogate-based Nonlinear Programming and Mixed-Integer Nonlinear Programming approaches are applied and compared in terms of computational efficiency and solution accuracy. Moreover, process performance results are examined by applying several data analytics techniques to gain insights into the material-process correlations. Data-driven classifiers and neural networks can accurately predict whether a material is likely to satisfy purity, recovery, and energy constraints when operated at optimal process conditions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Learning to simulate high energy particle collisions from unlabeled data

In many scientific fields which rely on statistical inference, simulations are often used to map from theoretical models to experimental data, allowing scientists to test model predictions against experimental results. Experimental data is often reconstructed from indirect measurements causing the aggregate transformation from theoretical models to experimental data to be poorly-described analytically. Instead, numerical simulations are used at great computational cost. We introduce Optimal-Transport-based Unfolding and Simulation (OTUS), a fast simulator based on unsupervised machine-learning that is capable of predicting experimental data from theoretical models. Without the aid of current simulation information, OTUS trains a probabilistic autoencoder to transform directly between theoretical models and experimental data. Identifying the probabilistic autoencoder’s latent space with the space of theoretical models causes the decoder network to become a fast, predictive simulator with the potential to replace current, computationally-costly simulators. Here, we provide proof-of-principle results on two particle physics examples, Z-boson and top-quark decays, but stress that OTUS can be widely applied to other fields.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Integrating in situ environmental covariates in an American lobster catch model to improve impact assessment

The installation and operation of floating offshore wind power is an integral component of societal transition to renewable energy generation where fixed bottom offshore wind is not possible. However, it will cause unique ecosystem changes. To disentangle the effects of offshore wind installations from the concurrent effects of climate change and the fishing practices on commercially significant resources, we must develop detailed characterizations of the resources before development occurs. In the Gulf of Maine, American lobster is the most commercially and culturally important fishery. At the time of writing, this is the largest fishery by value in North America. Our understanding of baseline localized parameters (such as catch per trap at the spatial scale of individual turbines) should be informed by relationships to environmental, biological, and survey-specific functional drivers of catch. A more mechanistic understanding of catch will allow for strategic adjustments to Post- Deployment fishery responses and ultimately, the development of research- and commercial-scale floating offshore wind development. Here, we used survey data from the New England Aqua Ventus Pre-Construction Commercial Trapping Survey to develop Generalized Additive Models describing seasonal catch per trap for legal and sublegal lobsters. We found fall catch to be nearly twice that of spring. Bottom temperature dynamics could be used to predict catch, and the Fall survey was associated with a warmer temperature regime. By using analytical tools that incorporate environmental heterogeneity, we developed monitoring methods from preconstruction baseline data that will be applicable over the post-construction operating period of an offshore wind farm.

BACI↗
DOE OSTI2020

Ion Solvation Engineering: How to Manipulate the Multiplicity of the Coordination Environment of Multivalent Ions

During free energy analysis of solvation structures of free divalent cations, their ion pairs, and neutral aggregates in low dielectric solvents reveals the multiplicity of thermodynamically stable cation solvation configurations and identifies the micro- and macroscopic factors responsible for this phenomenon. Specifically, we show the role of ion-solvent interactions and solvent mixtures in determining the cation solvation free energy landscapes. We show that it is the entropic contribution of solvent degrees of freedom that is responsible for the solvation multiplicity, and the mutual balance between enthalpic and entropic forces or their concerted contributions is what ultimately defines the most stable ion solvation configuration and creates new ones. We show general consequences of ion solvation multiplicity on thermodynamics of complex electrolytes, specifically in the context of homogeneous or interfacial charge transfer. Identified factors and their interplay provide a pathway to formulation of solvation design rules that can be used to control bulk solvation, interfacial chemistry, and charge transfer. Our findings also suggest experimentally testable predictions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Electron-Only Magnetic Reconnection and Inverse Magnetic-Energy Transfer at Subion Scales

We derive, and validate numerically, an analytical model for electron-only magnetic reconnection applicable to strongly magnetized plasmas. Our model predicts subion-scale reconnection rates significantly higher than those pertaining to large-scale reconnection, aligning with recent observations and simulations. Here, we apply this reconnection model to the problem of inverse magnetic energy transfer at subion scales. We derive time-dependent scaling laws for the magnetic energy decay and the typical magnetic structure dimensions that differ from those previously found in the magnetohydrodynamics regime. These scaling laws are validated via two- and three-dimensional simulations, demonstrating that subion-scale magnetic fields can reach large, system-size scales via successive coalescence.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

Electrical Submersible Pumps: A System Modeling Approach for Power Quality Analysis With Variable Frequency Drives

This article proposes a high-level engineering guide to develop an integrated system model for power quality analysis in electrical submersible pump (ESP) applications with variable frequency drives (VFDs). Such analyses are troublesome in the industry due to the complexity of these systems. In this article, simple steps to perform system integration analyses of such arrangements, including torsional analysis, are developed. A simplified VFD-ESP model suitable for coupled electrical and mechanical analysis in steady state is proposed in all their configurations of practical interest. Such a model can be easily implemented in common simulation software, significantly reducing engineering efforts for implementation and analysis. The focus of the proposed model is the prediction of ESP failures that might result from a poor power quality caused by VFDs. Analytical expressions of different types of harmonics in these systems, as well as their accurate locations in the frequency domain, including their interharmonics and common-mode harmonics, are derived for this purpose. Further, the effectiveness of the proposed model is verified through offline and real-time hybrid simulation results. Finally, a comparison between simulation results obtained using the proposed model and measurements collected on a down-scale laboratory prototype is carried out to demonstrate the accuracy of the suggested modeling approach.

42 ENGINEERING↗
DOE OSTI2020

Where do $AMS-02$ anti-helium events come from?

We discuss the origin of the anti-helium-3 and -4 events possibly detected by AMS-02. Using up-to-date semi-analytical tools, we show that spallation from primary hydrogen and helium nuclei onto the ISM predicts a anti-helium-3 flux typically one to two orders of magnitude below the sensitivity of AMS-02 after 5 years, and a anti-helium-4 flux roughly 5 orders of magnitude below the AMS-02 sensitivity. We argue that dark matter annihilations face similar difficulties in explaining this event. We then entertain the possibility that these events originate from anti-matter-dominated regions in the form of anti-clouds or anti-stars. In the case of anti-clouds, we show how the isotopic ratio of anti-helium nuclei might suggest that BBN has happened in an inhomogeneous manner, resulting in anti-regions with a anti-baryon-to-photon ratio of one in a thousand of the baryon-to-photon ratio. We discuss properties of these regions, as well as relevant constraints on the presence of anti-clouds in our Galaxy. We present constraints from the survival of anti-clouds in the Milky-Way and in the early Universe, as well as from gamma-ray observations. In particular, these require the anti-clouds to be almost free of normal matter. We also discuss an alternative where anti-domains are dominated by surviving anti-stars. We suggest that part of the unindentified sources in the 3FGL catalog can originate from anti-clouds or anti-stars. AMS-02 and GAPS data could further probe this scenario.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Multisensor Agile Adaptive Sampling of Convective Storms Driven by Real-time Analytics

Convective storms vertically transport water vapor and condensate from Earth’s surface to the upper troposphere. Life on Earth is fundamentally linked to this transport which determines the hydrological cycle, and the intensity of severe weather responsible for the destruction of life and property. Despite advances in high-resolution modeling and better observational capabilities, the scientific community continues to be confronted with knowledge gaps about convective storms that limit our predictive capabilities. The ongoing developments in the high-resolution Energy Exascale Earth System Model (E3SM), large eddy simulations, and AI-based analytics to evaluate uncertainties are expected to provide a comprehensive framework for new scientific discovery. The model-experiment (MODEX) approach suggests that the aforementioned advancements in model development and AI-based inference techniques should be complemented by similar advancements in the experimental (observational) side so that the former does not outstrip the ability of the latter to provide meaningful constraints. What are the recent advancements in observations that will provide the necessary leap forward in improving our predictive capabilities? To address this question, we propose a new experimental paradigm called Multisensor Agile Adaptive Sampling (MAAS) that capitalizes on advancements in communications (5G), computational resources (edge/fog computing), sensor capabilities, and machine learning (ML) and AI techniques (Kollias et al., 2020). The MAAS framework allows for the collection of higher spatiotemporal resolution and quality observations of convective storms than is traditionally possible. The MAAS framework is scalable and applicable to atmospheric observatories such as those operated by the Department of Energy (DoE) Atmospheric Radiation Measurement (ARM) facility.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Predicting battery capacity from impedance at varying temperature and state of charge using machine learning

Prediction of battery health from electrochemical impedance spectroscopy (EIS) data can enable rapid measurement of battery state in real-world applications without using additional sensors or time-consuming performance measurements. However, deconvoluting the effect of capacity, state of charge, and temperature on EIS response is complicated analytically. Here, various machine-learning models, such as linear, Gaussian process, random forest, and artificial neural network regression, are utilized to predict capacity from EIS using hundreds of capacity, direct current (DC) resistance, and EIS measurements recorded under varying conditions of health, temperature, and state of charge (SOC). Several feature extraction and selection methods from traditional electrochemical analysis and statistical modeling are explored using machine-learning pipelines. EIS data from just two frequencies can accurately predict capacity, and interrogation shows that the optimal set of frequencies is not usually intuitive. Best results are achieved with an ensemble model, which predicts battery capacity with a mean absolute error of 1.9% on data from unobserved cells.

25 ENERGY STORAGE↗
DOE OSTI2024

Simultaneously improving accuracy and computational cost under parametric constraints in materials property prediction tasks

Abstract Modern data mining techniques using machine learning (ML) and deep learning (DL) algorithms have been shown to excel in the regression-based task of materials property prediction using various materials representations. In an attempt to improve the predictive performance of the deep neural network model, researchers have tried to add more layers as well as develop new architectural components to create sophisticated and deep neural network models that can aid in the training process and improve the predictive ability of the final model. However, usually, these modifications require a lot of computational resources, thereby further increasing the already large model training time, which is often not feasible, thereby limiting usage for most researchers. In this paper, we study and propose a deep neural network framework for regression-based problems comprising of fully connected layers that can work with any numerical vector-based materials representations as model input. We present a novel deep regression neural network, iBRNet, with branched skip connections and multiple schedulers, which can reduce the number of parameters used to construct the model, improve the accuracy, and decrease the training time of the predictive model. We perform the model training using composition-based numerical vectors representing the elemental fractions of the respective materials and compare their performance against other traditional ML and several known DL architectures. Using multiple datasets with varying data sizes for training and testing, We show that the proposed iBRNet models outperform the state-of-the-art ML and DL models for all data sizes. We also show that the branched structure and usage of multiple schedulers lead to fewer parameters and faster model training time with better convergence than other neural networks. Scientific contribution: The combination of multiple callback functions in deep neural networks minimizes training time and maximizes accuracy in a controlled computational environment with parametric constraints for the task of materials property prediction.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Using Artificial Neural Networks to Predict Physical Properties of Membrane Polymers

Membrane polymers are a promising technology for use in many challenging gas separation applications. Here, the techniques of computer-aided molecular design can be used to search through the massive molecular space of heteropolymers and develop a set of likely candidate repeat units matching specific physical property targets. However, reasonably accurate property prediction algorithms are needed, but these algorithms must be very fast in order to be combined with an optimization framework. Artificial neural networks (ANNs), a branch of machine learning, are applied in this work to predict the physical properties of polymers. All of the physical properties investigated were found to be predicted by ANNs with R 2 scores exceeding 0.82.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Utilizing metal cation reactions with carbonyl sulfide to remove isobaric interferences in tandem inductively coupled plasma mass spectrometry analyses

Carbonyl sulfide (OCS) was used as a reaction gas to investigate gas phase metal (M + ) ion-molecule reactions using the Agilent 8900 inductively coupled plasma tandem mass spectrometer (ICP-MS/MS) to yield insight on how this gas may be used to remove isobaric interferences in analytical measurements. The experimental work was paired with density functional theory (DFT) calculations of the reaction enthalpy to predict whether M + will react with OCS. A multi-element standard containing 46 elements ranging from 9 to 208 u was analyzed in the presence and absence of OCS. When a reaction was observed, the dominant product was the sulfide (MS + ). Oxide products were also observed for many M + but formation was less efficient with OCS than previously observed with other reaction gases. This is likely due to the weaker OC-S bond that makes MS + formation more favorable. Increasing the flow rate from 0.1 to 0.2 mL/min (corresponding to a change in reaction gas pressure from 0.35 to 0.53 Pa (2.6 to 4.0 mTorr)) generally resulted in greater MS + production, including the secondary product MS 2 + for a few cations. The early lanthanide series ions (La + , Ce + , Pr + and Nd + ) produced greater quantities of MO + at the higher pressure, although MS + products were still the dominant product. The DFT-predicted reaction enthalpies were consistent with the observed sulfide formation, with an accuracy >90%; however, model predictions were less accurate for the minor and higher order products (< 77% for MO + ). Finally, the work presented here continues a systematic study of ion-molecule reactions in ICP-MS/MS to understand and develop new and novel ways to analyze complex mixtures with minimal pre-analysis treatment.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗