Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “predictive mapping”

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 883 records · Page 49
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Urban Seismic Site Characterization by Fiber-Optic Seismology

ccurate ground motion prediction requires detailed site effect assessment, but in urban areas where such assessments are most important, geotechnical surveys are difficult to perform, limiting their availability. Distributed acoustic sensing (DAS) offers an appealing alternative by repurposing existing fiber-optic cables, normally employed for telecommunication, as an array of seismic sensors. We present a proof-of-concept demonstration by using DAS to produce high-resolution maps of the shallow subsurface with the Stanford DAS array, California. We describe new methods and their assumptions to assess H/V spectral ratio—a technique widely used to estimate the natural frequency of the soil—and to extract Rayleigh wave dispersion curves from ambient seismic field. These measurements are jointly inverted to provide models of shallow seismic velocities and sediment thicknesses above bedrock in central campus. The good agreement with an independent survey validates the methodology and demonstrates the power of DAS for microzonation.

Geochemistry & Geophysics↗
DOE OSTI2026

Explainable tokamak-agnostic forecasting of fusion plasma instability via megahertz turbulent fluctuations

Scientific applications of artificial intelligence (AI) often remain limited by device-specific training and unexplained “black-box” approaches, creating fundamental barriers to cross-system generalization. This challenge is critical for nuclear fusion, where future reactors will have limited operational data for AI training. Here, we demonstrate that our neural network, trained solely on megahertz-scale turbulence measurements from one machine (DIII-D), forecasts Type-I edge localized mode (ELM) onsets in a different tokamak (KSTAR) through zero-shot weight transfer following physics-consistent preprocessing without device-specific retraining. Through an explainable AI framework combining gradient-weighted class activation mapping with physics validation, we reveal that our network can internalize physics relationships governing the ELM instabilities rather than memorizing device-specific patterns. The network perceives spatiotemporal features that correlate consistently with independently calculated instability growth rates, magnetohydrodynamic stability limits, and pedestal structure dynamics. Statistical analyses of dimensionally-reduced saliency features reveal the identical triangular features between the saliency representations, instability growth rates, and prediction probability across tokamaks, providing evidence that our forecasting system can show tokamak-agnostic generalization. This work contributes to a foundation for explainable scientific AI systems, where cross-system developments are essential for transcending traditional domain-specific constraints.

AI↗
DOE OSTI2026

Universal CMB 𝐵-mode spectrum from early causal tensor sources

Many early Universe scenarios predict postinflationary tensor perturbations from causality-limited, subhorizon sources. While the microphysical details may be different, as long as these sources are bounded in duration and correlation length, their tensor power spectra exhibit a universal scaling behavior at small wave number: 𝒫 ℎ ⁡(𝑘)∝𝑘 3 , corresponding to white noise on superhorizon scales at the time of production. If these early causal tensor sources (ECTs) exclusively produce gravitational waves before redshift 𝑧 ∼10 5 , this scaling is realized on all of the scales observed in the CMB, and thus yields a universal multipole distribution for the 𝐵-mode angular power spectrum. Unlike the scale-invariant distributions of inflationary 𝐵 modes, ECTs generically predict enhanced power on small scales and suppressed power on large scales, which allows these source classes to be distinguished given measurements over a sufficient range of angular scales. In this paper, we introduce a unified framework for characterizing ECTs and demonstrate how their universal infrared scaling manifests in low-frequency observables, including CMB 𝐵 modes and stochastic gravitational wave spectral densities. We illustrate this mapping with representative case studies of this universality class involving first-order phase transitions, topological defects, and enhanced scalar perturbations, which source tensor modes at second order in perturbation theory.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Optimizing transmit field inhomogeneity of parallel RF transmit design in 7T MRI using deep learning

Ultrahigh field (UHF) Magnetic Resonance Imaging (MRI) provides a higher signal-to-noise ratio and, thereby, higher spatial resolution. However, UHF MRI introduces challenges such as transmit radiofrequency (RF) field (B+1) inhomogeneities, leading to uneven flip angles and image intensity anomalies. These issues can significantly degrade imaging quality and its medical applications. This study addresses B+1 field homogeneity through a novel deep learning-based strategy. Traditional methods like Magnitude Least Squares (MLS) optimization have been effective but are time-consuming and dependent on the patient’s presence. Recent machine learning approaches, such as RF Shim Prediction by Iteratively Projected Ridge Regression and deep learning frameworks, have shown promise but face limitations like extensive training times and oversimplified architectures. We propose a two-step deep learning strategy. First, we obtain the desired reference RF shimming weights from multi-channel B+1 fields using random-initialized Adaptive Moment Estimation. Then, we employ Residual Networks (ResNets) to train a model that maps B+1 fields to target RF shimming outputs. Our approach does not rely on pre-calculated reference optimizations for the testing process and efficiently learns residual functions. Comparative studies with traditional MLS optimization demonstrate our method’s advantages in terms of speed and accuracy. The proposed strategy achieves a faster and more efficient RF shimming design, significantly improving imaging quality at UHF. This advancement holds potential for broader applications in medical imaging and diagnostics.

Lu, Zhengyi [Vanderbilt University]↗
DOE OSTI2025

White Paper: Scalable Digital Twin Capabilities for Aging and Surveillance of Engineered Systems

This white paper presents a multi-year initiative to develop practical, secure, and scalable digital twin capabilities for engineered systems in aging and surveillance contexts—an approach pioneered at the National Nuclear Security Administration (NNSA) Lawrence Livermore National Laboratory (LLNL) that maps directly onto the needs and ambitions of the Navy for ship- and fleet-level digital twins. LLNL’s work in building part- and process-level digital twins for advanced manufacturing, with a vision to scale up to entire factory floors and, ultimately, enterprise-wide digital twins, offers an adaptable pathway for the Navy as it seeks to modernize lifecycle management, readiness, and predictive maintenance across ships and fleets. For our application, we integrate physics-based modeling with automated data ingestion, processing, and AI-driven calibration, creating hybrid models that are both interpretable and data responsive. We modernized legacy workflows, established centralized data infrastructure, automated experimental pipelines, and demonstrated end-to-end coupling of accelerated aging data with finite element simulations via optimization and surrogate modeling. The result is a generalizable framework that supports part-level digital twins today and lays the groundwork for future system-level twins suitable for Navy applications.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Global 10 m Land Use Land Cover Datasets: A Comparison of Dynamic World, World Cover and Esri Land Cover

The European Space Agency’s Sentinel satellites have laid the foundation for global land use land cover (LULC) mapping with unprecedented detail at 10 m resolution. We present a cross-comparison and accuracy assessment of Google’s Dynamic World (DW), ESA’s World Cover (WC) and Esri’s Land Cover (Esri) products for the first time in order to inform the adoption and application of these maps going forward. For the year 2020, the three global LULC maps show strong spatial correspondence (i.e., near-equal area estimates) for water, built area, trees and crop LULC classes. However, relative to one another, WC is biased towards over-estimating grass cover, Esri towards shrub and scrub cover and DW towards snow and ice. Using global ground truth data with a minimum mapping unit of 250 m 2 , we found that Esri had the highest overall accuracy (75%) compared to DW (72%) and WC (65%). Across all global maps, water was the most accurately mapped class (92%), followed by built area (83%), tree cover (81%) and crops (78%), particularly in biomes characterized by temperate and boreal forests. The classes with the lowest accuracies, particularly in the tundra biome, included shrub and scrub (47%), grass (34%), bare ground (57%) and flooded vegetation (53%). When using European ground truth data from LUCAS (Land Use/Cover Area Frame Survey) with a minimum mapping unit of <100 m 2 , we found that WC had the highest accuracy (71%) compared to DW (66%) and Esri (63%), highlighting the ability of WC to resolve landscape elements with more detail compared to DW and Esri. Although not analyzed in our study, we discuss the relative advantages of DW due to its frequent and near real-time data delivery of both categorical predictions and class probability scores. We recommend that the use of global LULC products should involve critical evaluation of their suitability with respect to the application purpose, such as aggregate changes in ecosystem accounting versus site-specific change detection in monitoring, considering trade-offs between thematic resolution, global versus. local accuracy, class-specific biases and whether change analysis is necessary. We also emphasize the importance of not estimating areas from pixel-counting alone but adopting best practices in design-based inference and area estimation that quantify uncertainty for a given study area.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

A Geo-Data Science Method for Assessing Unconventional Rare-Earth Element Resources in Sedimentary Systems

Abstract Rare-earth elements (REEs) supply raw materials that constitute many of our modern critical infrastructure, defense, technology, and electrification needs. Despite REE accumulations occurring in conventional bedrock and ion-adsorption deposits sourced from weathering of igneous rocks, unconventional host materials such as coal and related sedimentary strata have been identified as promising sources of REEs to meet growing demand. To maximize the potential of unconventional resources such as REE-coal systems, new approaches are needed overcome challenges from mineral systems with no known deposits and areas with sparse geochemical data. This article presents a systematic knowledge-data resource assessment method for predicting and identifying REE resource potential and occurrence in these unconventional systems. The method utilizes a geologic and geospatial knowledge-data approach informed and guided by REE accumulation mechanisms to systematically assess and identify areas of higher enrichment. An assessment of the Powder River Basin is presented as a test case to demonstrate the method workflow and results. The key output is a potential enrichment score map reported with varying confidence levels based on the amount of supporting evidence. Results from the test case indicate several locations with promising potential for different types of coal-REE deposits, demonstrating the viability of the method for exploration and assessment of unconventional REE resources. The method is flexible by design and, with sufficient applicable knowledge and data, can be adapted for assessing critical mineral systems in other sedimentary systems as well.

58 GEOSCIENCES↗
DOE OSTI2024

Scientific Foundations and Approaches for Qualification of Additively Manufactured Structural Components

Additive manufacturing (AM) maintains a wide process window that enables complex designs otherwise unattainable via conventional production technologies. However, the lack of confidence in qualifying AM parts that leverage AM process–structure–property–performance (PSPP) relationships stymies design optimization and adoption of AM. While continuing efforts to map fundamental PSPP relationships that cover the potential design space, we first need pragmatic and then long-term solutions that overcome challenges associated with qualifying AM-designed parts. Two pragmatic solutions include: (1) AM material specifications to substantiate process reproducibility, and (2) component risk categorization to associate system risk relative to part performance and required part quality. A novel qualification paradigm under development involves efficient prediction of part performance over wide-ranging PSPP relationships through targeted testing and computational simulation. Here this paper describes projects at Sandia National Laboratories on PSPP relationship discovery, these pragmatic approaches, and the novel qualification approach.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Do chaotic field lines cause fast reconnection in coronal loops?

Over the past decade, Boozer has argued that three-dimensional (3D) magnetic reconnection fundamentally differs from two-dimensional reconnection due to the fact that the separation between any pair of neighboring field lines almost always increases exponentially over distance in a 3D magnetic field. According to Boozer, this feature makes 3D field-line mapping chaotic and exponentially sensitive to small non-ideal effects; consequently, 3D reconnection can occur without intense current sheets. We test Boozer's theory via ideal and resistive reduced magnetohydrodynamic simulations of the Boozer–Elder coronal loop model driven by sub-Alfvénic footpoint motions [A. H. Boozer and T. Elder, Phys. Plasmas 28, 062303 (2021)]. In this work, our simulation results significantly differ from their predictions. The ideal simulation shows that Boozer and Elder under-predict the intensity of current density due to missing terms in their reduced model equations. Furthermore, resistive simulations of varying Lundquist numbers show that the maximal current density scales linearly rather than logarithmically with the Lundquist number.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

Coupled Monte Carlo Transport and Conjugate Heat Transfer for Wire-Wrapped Bundles Within the MOOSE Framework

Cardinal is an open-source application that couples OpenMC Monte Carlo transport and NekRS computational fluid dynamics (CFD) to the Multiphysics Object-Oriented Simulation Environment (MOOSE), closing neutronics and thermal-fluid gaps in conducting high-resolution multiscale and multiphysics analyses of nuclear systems. We first provide a brief introduction to Cardinal's software design, data mapping, and coupling strategy to highlight our approach to overcoming common challenges in high-fidelity multiphysics simulations. Here we then present two Cardinal simulations for hexagonal pin bundles. The first is a validation of Cardinal's conjugate heat transfer coupling of NekRS's Reynolds-Averaged Navier Stokes model with MOOSE's heat conduction physics for a bare seven-pin Freon-12 bundle flow experiment. Predictions for pin surface temperatures under three different heating modes agree reasonably well with experimental data and similar CFD modeling from the literature. The second simulation is a multiphysics coupling of OpenMC, NekRS, and BISON for a reduced-scale, seven-pin wire-wrapped version of an Advanced Burner Reactor bundle. Wire wraps are approximated using a momentum source model, and coupled predictions are provided for velocity, temperature, and power distribution.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2023

Improved quality metrics for association and reproducibility in chromatin accessibility data using mutual information

Correlation metrics are widely utilized in genomics analysis and often implemented with little regard to assumptions of normality, homoscedasticity, and independence of values. This is especially true when comparing values between replicated sequencing experiments that probe chromatin accessibility, such as assays for transposase-accessible chromatin via sequencing (ATAC-seq). Such data can possess several regions across the human genome with little to no sequencing depth and are thus non-normal with a large portion of zero values. Despite distributed use in the epigenomics field, few studies have evaluated and benchmarked how correlation and association statistics behave across ATAC-seq experiments with known differences or the effects of removing specific outliers from the data. Here, we developed a computational simulation of ATAC-seq data to elucidate the behavior of correlation statistics and to compare their accuracy under set conditions of reproducibility. Using these simulations, we monitored the behavior of several correlation statistics, including the Pearson’s R and Spearman’s ρ coefficients as well as Kendall’s τ and Top–Down correlation. We also test the behavior of association measures, including the coefficient of determination R 2 , Kendall’s W, and normalized mutual information. Our experiments reveal an insensitivity of most statistics, including Spear man’s ρ, Kendall’s τ, and Kendall’s W, to increasing differences between simulated ATAC-seq replicates. The removal of co-zeros (regions lacking mapped sequenced reads) between simulated experiments greatly improves the estimates of correlation and association. After removing co-zeros, the R 2 coefficient and normalized mutual information display the best performance, having a closer one-to-one relationship with the known portion of shared, enhanced loci between simulated replicates. When comparing values between experimental ATAC-seq data using a random forest model, mutual information best predicts ATAC-seq replicate relationships. Collectively, this study demonstrates how measures of correlation and association can behave in epigenomics experiments. We provide improved strategies for quantifying relationships in these increasingly prevalent and important chromatin accessibility assays.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Mass Transport Limitations and Kinetic Consequences of Corn Stover Deacetylation

Alkaline pretreatment of herbaceous feedstocks such as corn stover prior to mechanical refining and enzymatic saccharification improves downstream sugar yields by removing acetyl moieties from hemicellulose. However, the relationship between transport phenomena and deacetylation kinetics is virtually unknown for such feedstocks and this pretreatment process. Here, we report the development of an experimentally validated reaction–diffusion model for the deacetylation of corn stover. A tissue-specific transport model is used to estimate transport-independent kinetic rate constants for the reactive extraction of acetate, hemicellulose and lignin from corn stover under representative alkaline conditions (5–7 g L -1 NaOH, 10 wt% solids loadings) and at low to mild temperatures (4–70°C) selected to attenuate individual component extraction rates under differential kinetic regimes. The underlying transport model is based on microstructural characteristics of corn stover derived from statistically meaningful geometric particle and pore measurements. These physical descriptors are incorporated into distinct particle models of the three major anatomical fractions (cobs, husks and stalks) alongside an unsorted, aggregate corn stover particle, capturing average Feret lengths of 917–1239 μm and length-to-width aspect ratios of 1.8–2.9 for this highly heterogeneous feedstock. Individual reaction–diffusion models and their resulting particle model ensembles are used to validate and predict anatomically-specific and bulk feedstock performance under kinetic-controlled vs. diffusion-controlled regimes. In general, deacetylation kinetics and mass transfer processes are predicted to compete on similar time and length scales, emphasizing the significance of intraparticle transport phenomena. Critically, we predict that typical corn stover particles as small as ~2.3 mm in length are entirely diffusion-limited for acetate extraction, with experimental effectiveness factors calculated to be 0.50 for such processes. Debilitatingly low effectiveness factors of 0.021–0.054 are uncovered for cobs—implying that intraparticle mass transfer resistances may impair observable kinetic measurements of this anatomical fraction by up to 98%. These first-reported quantitative maps of reaction vs. diffusion control link fundamental insights into corn stover anatomy, biopolymer composition, practical size reduction thresholds and their kinetic consequences. These results offer a guidepost for industrial deacetylation reactor design, scale-up and feedstock selection, further establishing deacetylation as a viable biorefinery pretreatment for the conversion of lignocellulosics into value-added fuels and chemicals.

09 BIOMASS FUELS↗
DOE OSTI2025

Constraints on cosmology and baryonic feedback with joint analysis of Dark Energy Survey Year 3 lensing data and ACT DR6 thermal Sunyaev-Zel'dovich effect observations

We present a joint analysis of weak gravitational lensing (shear) data obtained from the first three years of observations by the Dark Energy Survey and thermal Sunyaev-Zel'dovich (tSZ) effect measurements from a combination of Atacama Cosmology Telescope (ACT) and Planck data. A combined analysis of shear (which traces the projected mass) with the tSZ effect (which traces the projected gas pressure) can jointly probe both the distribution of matter and the thermodynamic state of the gas, accounting for the correlated effects of baryonic feedback on both observables. We detect the shear$~\times~$tSZ cross-correlation at a 21$\sigma$ significance, the highest to date, after minimizing the bias from cosmic infrared background leakage in the tSZ map. By jointly modeling the small-scale shear auto-correlation and the shear$~\times~$tSZ cross-correlation, we obtain $S_8 = 0.811^{+0.015}_{-0.012}$ and $\Omega_{\rm m} = 0.263^{+0.023}_{-0.030}$, results consistent with primary CMB analyses from Planck and P-ACT. We find evidence for reduced thermal gas pressure in dark matter halos with masses $M < 10^{14} \, M_{\odot}/h$, supporting predictions of enhanced feedback from active galactic nuclei on gas thermodynamics. A comparison of the inferred matter power suppression reveals a $2-4\sigma$ tension with hydrodynamical simulations that implement mild baryonic feedback, as our constraints prefer a stronger suppression. Finally, we investigate biases from cosmic infrared background leakage in the tSZ-shear cross-correlation measurements, employing mitigation techniques to ensure a robust inference. Our code is publicly available on GitHub.

Pandey, S. [Johns Hopkins U.; Columbia U.] (ORCID:↗
DOE OSTI2022

Hyperspectral remote sensing-based plant community map for region around NGEE-Arctic intensive research watersheds at Seward Peninsula, Alaska, 2017-2019

Using airborne hyperspectral remote sensing data from NASA Airborne Visible-Infrared Imaging Spectrometer- Next Generation (AVIRIS-NG) platforms in a region near NGEE-Arctic intensive watersheds at Seward peninsula of Alaska, high resolution (5m) maps of plant community distribution were developed and included in this data collected. AVIRIS-NG data collected over 2017-2019 period were used to develop deep neural networks, trained using vegetation plot observations collected at NGEE-Arctic watersheds at Kougarok, Council and Teller. A hierarchical vegetation classification scheme consisting of six classes at Level I, and 16 classes at Level II contained in two .txt files were used to developed the plant community maps for the region. Two geospatial raster data files (.tif) at both thematic levels are shared in this data collection. Data files in this collection use Alaska Albers Equal Area projection. Readme files available in three formats (*.html, *.md, *.pdf) and one *.png visualization map.The Next-Generation Ecosystem Experiments: Arctic (NGEE Arctic), was a research effort to reduce uncertainty in Earth System Models by developing a predictive understanding of carbon-rich Arctic ecosystems and feedbacks to climate. NGEE Arctic was supported by the Department of Energy's Office of Biological and Environmental Research.The NGEE Arctic project had two field research sites: 1) located within the Arctic polygonal tundra coastal region on the Barrow Environmental Observatory (BEO) and the North Slope near Utqiagvik (Barrow), Alaska and 2) multiple areas on the discontinuous permafrost region of the Seward Peninsula north of Nome, Alaska.Through observations, experiments, and synthesis with existing datasets, NGEE Arctic provided an enhanced knowledge base for multi-scale modeling and contributed to improved process representation at global pan-Arctic scales within the Department of Energy's Earth system Model (the Energy Exascale Earth System Model, or E3SM), and specifically within the E3SM Land Model component (ELM).

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Protein model quality assessment using rotation–equivariant transformations on point clouds

Machine learning research concerning protein structure has seen a surge in popularity over the last years with promising advances for basic science and drug discovery. Working with macromolecular structure in a machine learning context requires an adequate numerical representation, and researchers have extensively studied representations such as graphs, discretized 3D grids, and distance maps. As part of CASP14, we explored a new and conceptually simple representation in a blind experiment: atoms as points in 3D, each with associated features. These features—initially just the basic element type of each atom—are updated through a series of neural network layers featuring rotation-equivariant convolutions. Starting from all atoms, we further aggregate information at the level of alpha carbons before making a prediction at the level of the entire protein structure. We find that this approach yields competitive results in protein model quality assessment despite its simplicity and despite the fact that it incorporates minimal prior information and is trained on relatively little data. As a result, its performance and generality are particularly noteworthy in an era where highly complex, customized machine learning methods such as AlphaFold 2 have come to dominate protein structure prediction.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Emergence of distinct electronic states in epitaxially-fused PbSe quantum dot superlattices

Abstract Quantum coupling in arrayed nanostructures can produce novel mesoscale properties such as electronic minibands to improve the performance of optoelectronic devices, including ultra-efficient solar cells and infrared photodetectors. Colloidal PbSe quantum dots (QDs) that self-assemble into epitaxially-fused superlattices (epi-SLs) are predicted to exhibit such collective phenomena. Here, we show the emergence of distinct local electronic states induced by crystalline necks that connect individual PbSe QDs and modulate the bandgap energy across the epi-SL. Multi-probe scanning tunneling spectroscopy shows bandgap modulation from 0.7 eV in the QDs to 1.1 eV at their necks. Complementary monochromated electron energy-loss spectroscopy demonstrates bandgap modulation in spectral mapping, confirming the presence of these distinct energy states from necking. The results show the modification of the electronic structure of a precision-made nanoscale superlattice, which may be leveraged in new optoelectronic applications.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Velocity Dispersion σ aper Aperture Corrections as a Function of Galaxy Properties from Integral-field Stellar Kinematics of 10,000 MaNGA Galaxies

Abstract The second moment of the stellar velocity within the effective radius, denoted by σ e 2 , is a crucial quantity in galaxy studies, as it provides insight into galaxy properties and their mass distributions. However, large spectroscopic surveys typically do not measure σ e directly, instead providing σ aper , the second moment of the stellar velocity within a fixed fiber aperture. In this paper, we derive an empirical aperture correction formula, given by σ aper / σ e = ( R aper / R e ) α , using spatially resolved stellar kinematics extracted from approximately 10,000 Sloan Digital Sky Survey–Mapping Nearby Galaxies at Apache Point Observatory integral field unit observations. Our analysis reveals a strong dependence of α on the r -band absolute magnitude M r , g – i color, and Sérsic index n Ser , where α values are lower for brighter, redder galaxies with higher Sérsic indices. Our results demonstrate that the aperture correction derived from previous literature on early-type galaxies cannot be applied to predict the aperture corrections for galaxies with intermediate Sérsic indices. We provide a lookup table of α values for different galaxy types, with parameters in the ranges of −18 > M r > −24, 0.4 < g – i < 1.6, and 0 < n Ser < 8. A Python script is provided to obtain the correction factors from the lookup table.

Astronomy & Astrophysics↗
DOE OSTI2024

Mixed Delay/Nondelay Embeddings Based Neuromorphic Computing with Patterned Nanomagnet Arrays

Patterned nanomagnet arrays (PNAs) have been shown to exhibit a strong geometrically frustrated dipole interaction. Some PNAs have also shown emergent domain wall dynamics. Previous works have demonstrated methods to physically probe these magnetization dynamics of PNAs to realize neuromorphic reservoir systems that exhibit chaotic dynamical behavior and high-dimensional nonlinearity. These PNA reservoir systems from prior works leverage echo state properties and linear/nonlinear short-term memory of component reservoir nodes to map and preserve the dynamical information of the input time-series data into nondelay spatial embeddings. Such mappings enable these PNA reservoir systems to imitate and predict/forecast the input time series data. However, these prior PNA reservoir systems are based solely on the nondelay spatial embeddings obtained at component reservoir nodes. As a result, they require a massive number of component reservoir nodes, or a very large spatial embedding (i.e., high-dimensional spatial embedding) per reservoir node, or both, to achieve acceptable imitation and prediction accuracy. These requirements reduce the practical feasibility of such PNA reservoir systems. To address this shortcoming, we present a mixed delay/nondelay embeddings-based PNA reservoir system. Our system uses a single PNA reservoir node with the ability to obtain a mixture of delay/nondelay embeddings of the dynamical information of the time-series data applied at the input of a single PNA reservoir node. Our analysis shows that when these mixed delay/nondelay embeddings are used to train a perceptron at the output layer, our reservoir system outperforms existing PNA-based reservoir systems for the imitation of NARMA 2, NARMA 5, NARMA 7, and NARMA 10 time series data, and for the short-term and long-term prediction of the Mackey Glass time series data.

Ti, Changpeng↗