SEARCH · Engineering Papers

Results for “forward modeling”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

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

At least 757 records · Page 42

All sources NASA NTRS DOE OSTI

DOE OSTI2021

A physics-informed and hierarchically regularized data-driven model for predicting fluid flow through porous media

This paper presents a new deep learning data-driven model for predicting structure dependent pore-fluid velocity fields in rock. The model is based on a Convolutional Auto-Encoder (CAE) artificial neural network capable of learning from image data generated by direct numerical simulations of fluid flow through pore-structures, such as by Lattice Boltzmann or molecular dynamics methods. The main novelty of the model in comparison to previous CAE-based data-driven approaches consists of three parts. The first is a methodology for decomposing the full-domain of the porous media into sub-regions, or “sub-domains”, in order to reduce the overall size of the CAE, batch process the sub-domains in parallel, and enable the CAE to learn local and generalizable nonlinear mappings of pore-fluid velocities. The second consists of embedding the finite difference solutions of the incompressible Navier-Stokes and continuity equations into convolutional layers prior to the CAE in order to provide the CAE with knowledge of fluid dynamics physics (PhyFlow). The third main novelty is that the training of the CAE is regularized with a hierarchical loss function that encourages the learning of fluid flow patterns (in a way similar to ranked modes in principal component analysis), ranking from most to least important. This is shown to increase the stability in learning, reduce over-fitting, and promote interpretability of the CAE neural network layers (HierCAE). The comprehensive new data-driven model, which we call the PhyFlow-HierCAE model, is shown to exhibit improved accuracy and generalizability of flow field predictions over conventional CAE models, attributable to the embedded physical knowledge and the hierarchical regularization, as well as realize orders of magnitude speed-ups in computation times as a surrogate for the direct numerical simulations. Examples of training and forward predictions on unseen pore-structures are provided and evaluated for data from Lattice Boltzmann and molecular dynamics simulations of pore-fluid flow. The model is shown to be a fast and accurate emulator (or “surrogate”) for predicting effective permeability of unseen pore-structures based on learning from relatively small direct numerical simulation datasets.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2026

First Search for Dark Sector e+e- Explanations of the MiniBooNE Anomaly at MicroBooNE

The MiniBooNE anomaly has puzzled the neutrino physics community for over two decades. MicroBooNE has previously searched for anomalous electron and photon excesses as potential explanations but the mystery persists. In this poster, we present the first direct test of dark sector models in which dark neutrinos scatter, then decay into e+e− pairs and missing energy, mimicking the MiniBooNE signal. These models have recently emerged as compelling candidates that evade existing constraints. Using 6.87×1020 protons-on-target in the Booster Neutrino Beam, we select forward-going, coherently produced e+e− events in the MicroBooNE detector. We observe 95 events against a constrained background prediction of 69.7±17.3. We set the world's first direct limits on this class of dark sector models, excluding at 95% CL the majority of parameter space viable as a solution to the MiniBooNE anomaly.

Basque, Vincent Marc [Edinburgh U.] (ORCID:0000000↗

DOE OSTI2025

Tightest Mixed-Integer Programming Formulations for Quadratic SCUC Optimization

In this project, we developed new, tighter Mixed-Integer Programming (MIP) formulations for the combined Alternating Current (AC) Security-Constrained Unit Commitment (SCUC) and Security-Constrained Optimal Power Flow (SCOPF). The work addresses a critical challenge in power system operations: efficiently determining which generation units to commit and how to optimally dispatch them while maintaining network reliability constraints for both normal and contingency scenarios. Our efforts: 1. Advance the Understanding of SCUC/SCOPF Modeling: By introducing tighter MIP formulations and leveraging cutting-edge optimization tools (Julia/JuMP, PowerModels.jl), this project has pushed forward the state of the art in efficient power systems scheduling. 2. Enhance Technical and Economic Feasibility: The methods developed provide more accurate and potentially faster solutions to large-scale, realistic scheduling and dispatch problems in electric power systems, which can translate into improved reliability and potentially lower costs for grid operations. 3. Benefit to the Public: Greater efficiency in power system operations leads to cost savings for utilities and end-users. Improved reliability and integration of advanced modeling approaches can facilitate the adoption of clean energy resources and better accommodate uncertainties in renewable generation. Because this technology could impact bulk power markets and reliability, these innovations have far-reaching public benefits in terms of cost savings, reliability, and sustainability.

24 POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2020

Multizone Residential HVAC Modeling

This report will evaluate several building physics modeling tools to determine which are best suited for developing next-generation zoning technology. These modeling tools will be evaluated for their ability to predict room-level thermal loads, HVAC system operation, temperature and comfort predictions, and scalability. Any gaps in the existing tools will be identified and a path forward will be discussed.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

DOE OSTI2021

New method for detecting fast neutrino flavor conversions in core-collapse supernova models with two-moment neutrino transport

Fast-pairwise neutrino oscillations potentially affect many aspects of core-collapse supernovae (CCSNe): the explosion mechanism, neutrino signals, and nucleosynthesis in the ejecta. This particular mode of collective neutrino oscillations has a deep connection to the angular structure of neutrinos in momentum space; for instance, the appearance of electron-neutrino lepton number (ELN) angular crossings in momentum space is a good indicator of the occurrence of flavor conversions. However, many multidimensional CCSN simulations are carried out with approximate neutrino transport (such as two-moment methods), which limits access to the angular distributions of neutrinos, i.e., inhibits ELNcrossing searches. In this paper, we develop a new method of searching for ELN crossing in these CCSN simulations. The required data is the zeroth and first angular moments of neutrinos and the matter profile, all of which are available in CCSN models with the two-moment method. One of the novelties of our new method is the use of a ray-tracing neutrino transport to determine ELNs in the direction of the stellar center. It is designed to compensate for shortcomings of the crossing searches with only two angular moments. We assess the capability of the method by carrying out a detailed comparison with results of full Boltzmann neutrino transport in 1D and 2D CCSN models. In this work, we find that the ray-tracing neutrino transport improves the accuracy of crossing searches; indeed, the appearance/disappearance of the crossings is accurately detected even in the region of forward-peaked angular distributions. The new method is computationally cheap and has the benefit of efficient parallelization; hence, it will be useful for ELN-crossing searches in any CCSN models that employ two-moment neutrino transport.

79 ASTRONOMY AND ASTROPHYSICS↗

DOE OSTI2024

Perspectives on AI Architectures and Codesign for Earth System Predictability

Abstract Recently, the U.S. Department of Energy (DOE), Office of Science, Biological and Environmental Research (BER), and Advanced Scientific Computing Research (ASCR) programs organized and held the Artificial Intelligence for Earth System Predictability (AI4ESP) workshop series. From this workshop, a critical conclusion that the DOE BER and ASCR community came to is the requirement to develop a new paradigm for Earth system predictability focused on enabling artificial intelligence (AI) across the field, laboratory, modeling, and analysis activities, called model experimentation (ModEx). BER’s ModEx is an iterative approach that enables process models to generate hypotheses. The developed hypotheses inform field and laboratory efforts to collect measurement and observation data, which are subsequently used to parameterize, drive, and test model (e.g., process based) predictions. A total of 17 technical sessions were held in this AI4ESP workshop series. This paper discusses the topic of the AI Architectures and Codesign session and associated outcomes. The AI Architectures and Codesign session included two invited talks, two plenary discussion panels, and three breakout rooms that covered specific topics, including 1) DOE high-performance computing (HPC) systems, 2) cloud HPC systems, and 3) edge computing and Internet of Things (IoT). We also provide forward-looking ideas and perspectives on potential research in this codesign area that can be achieved by synergies with the other 16 session topics. These ideas include topics such as 1) reimagining codesign, 2) data acquisition to distribution, 3) heterogeneous HPC solutions for integration of AI/ML and other data analytics like uncertainty quantification with Earth system modeling and simulation, and 4) AI-enabled sensor integration into Earth system measurements and observations. Such perspectives are a distinguishing aspect of this paper. Significance Statement This study aims to provide perspectives on AI architectures and codesign approaches for Earth system predictability. Such visionary perspectives are essential because AI-enabled model-data integration has shown promise in improving predictions associated with climate change, perturbations, and extreme events. Our forward-looking ideas guide what is next in codesign to enhance Earth system models, observations, and theory using state-of-the-art and futuristic computational infrastructure.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2021

Keras2c: A library for converting Keras neural networks to real-time compatible C

With the growth of machine learning models and neural networks in measurement and control systems comes the need to deploy these models in a way that is compatible with existing systems. Existing options for deploying neural networks either introduce very high latency, require expensive and time consuming work to integrate into existing code bases, or only support a very limited subset of model types. We have therefore developed a new method called Keras2c, which is a simple library for converting Keras/TensorFlow neural network models into real-time compatible C code. It supports a wide range of Keras layers and model types including multidimensional convolutions, recurrent layers, multi-input/output models, and shared layers. Keras2c re-implements the core components of Keras/TensorFlow required for predictive forward passes through neural networks in pure C, relying only on standard library functions considered safe for real-time use. The core functionality consists of ~ 1500 lines of code, making it lightweight and easy to integrate into existing codebases. Keras2c has been successfully tested in experiments and is currently in use on the plasma control system at the DIII-D National Fusion Facility at General Atomics in San Diego.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2025

HexWeather: Hexagonal Spatial Data Aggregation for Weather-Driven Grid Resilience Analysis

Extreme weather accounts for over 8 0 % of major U.S. power outages since 2000, highlighting the need for spatial tools that align weather data with the irregular boundaries of electric infrastructure. This paper introduces HexWeather, a modular, resolution-aware framework for aggregating historical and forecasted weather data using Uber's H3 hexagonal spatial indexing system. Unlike traditional methods that rely on state or county-level grids, HexWeather enables weather analysis across custom geographies such as utility service areas where public datasets are often unavailable or misaligned. Using Open-Meteo data, we evaluate how H3 resolution affects anomaly detection, spatial variability, and forecast uncertainty across three scales: state, county, and utility. Results show that while coarse resolutions suffice for broad trend tracking, finer resolutions are essential for identifying localized variability and operational risks. By applying metrics like Z-score standard deviation and interquartile range, HexWeather quantifies the spatial spread of both historical anomalies and forecasted conditions, allowing users to assess resolution adequacy for each analysis. This framework supports rapid weather data reuse, reproducible anomaly detection, and predictive modeling for infrastructure resilience. By bridging spatial misalignment in traditional datasets and enabling retrospective and forward-looking analysis within the same pipeline, HexWeather lays the groundwork for better post event analysis, outage prediction, and resilience planning.

Morris, Jacob [ORNL]↗

DOE OSTI2023

Compilation of a Solar Mirror Materials Database and an Analysis of Natural and Accelerated Mirror Exposure and Degradation

The National Renewable Energy Laboratory (NREL) has been conducting exposure experiments on solar reflectors for over four decades. Thousands of mirror samples from over one hundred suppliers have been exposed to and monitored in a range of relevant environments. These test conditions include outdoor test settings and several controlled laboratory environments. These samples have been rigorously individually characterized using a series of reflectance measurements, visual inspections, and in some cases, in-depth composition analysis to identify degradation modes, reflectance losses, and other mirror properties integral to understanding the solar reflector's life cycle. Here, this paper compiles the decades of measurement data into a concise statistical analysis. It includes exposure and degradation data for numerous reflector types, including secondary-surface reflector permutations of polymer and glass superstrates with silver and aluminum reflectors as well as front-surface reflectors. The results herein are intended to analyze environmental stressors and degradation trends among various historical and state-of-the-art solar reflectors. It may be used to support solar reflector design, effective testing methodology, and inform manufacturing decisions moving forward. Presented are the results of the compiled database and an initial analysis for degradation rate modeling using full-spectrum and wavelength-dependent approaches. The database is a growing resource hosted on a live, publicly accessible website. In conjunction with the analysis presented here, it provides a valuable resource to the solar reflector manufacturing and testing industry.

14 SOLAR ENERGY↗

DOE OSTI2022

graphenv: a Python library for reinforcement learning on graph search spaces

Many important and challenging problems in combinatorial optimization (CO) can be expressed as graph search problems, in which graph vertices represent full or partial solutions and edges represent decisions that connect them. Graph structure not only introduces strong relational inductive biases for learning (Battaglia et al., 2018) - in this context, by providing a way to explicitly model the value of transitioning (along edges) between one search state (vertex) and the next - but lends itself to problems both with and without clearly defined algebraic structure. For example, classic CO problems on graphs such as the Traveling Salesman Problem (TSP) can be expressed as either pure graph search or integer programs. Other problems, however, such as molecular optimization, do no have concise algebraic formulations and yet are readily implemented as a graph search (V. et al., 2022; Zhou et al., 2019). Such "model-free" problems constitute a large fraction of modern reinforcement learning (RL) research owing to the fact that it is often much easier to write a forward simulation that expresses all of the state transitions and rewards, than to write down the precise mathematical expression of the full optimization problem. In the case of molecular optimization, for example, one can use domain knowledge alongside existing software libraries to model the effect of adding a single bond or atom to an existing but incomplete molecule, and let the RL algorithm build a model of how good a given decision is by "experiencing" the simulated environment many times through. In contrast, a model-based mathematical formulation that fully expresses all the chemical and physical constraints is intractable. In recent years, RL has emerged as an effective paradigm for optimizing searches over graphs and led to state-of-the-art heuristics for games like Go and chess, as well as for classical CO problems such as the TSP. This combination of graph search and RL, while powerful, requires non-trivial software to execute, especially when combining advanced state representations such as Graph Neural Networks (GNN) with scalable RL algorithms.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Transitioning GlideinWMS, a multi domain distributed workload manager, from GSI proxies to tokens and other granular credentials

GlideinWMS is a distributed workload manager that has been used in production for many years to provision resources for experiments like CERN’s CMS, many Neutrino experiments, and the OSG. Its security model was based mainly on GSI (Grid Security Infrastructure), using X.509 certificate proxies and VOMS (Virtual Organization Membership Service) extensions. Even when other credentials, like SSH keys, were used to authenticate with resources, proxies were also added all the time, to establish the identity of the requestor and the associated memberships or privileges. This single credential was used for everything and was, often implicitly, forwarded wherever needed. The addition of identity and access tokens and the phase-out of GSI forced us to reconsider the security model of GlideinWMS, to handle multiple credentials which can differ in type, technology, and functionality. Both identity tokens and access tokens are supported. GSI proxies even if no more mandatory, are still used, together with various JWT (JSON Web Token) based tokens and other certificates. The functionality of the credentials, defined by issuer, audience, and scope, also differ: a credential can allow access to a computing resource, or can protect the GlideinWMS framework from tampering, or can grant read or write access to storage, can provide an identity for accounting or auditing, or can provide a combination of any the formers. Furthermore, the tools in use do not include automatic forwarding and renewal of the new credentials so credential lifetime and renewal requirements became part of the discussion as well. In this paper, we will present how GlideinWMS was able to change its design and code to respond to all these changes.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

First Double-Differential Cross Section Measurement of Neutral-Current 𝜋 0 Production in Neutrino-Argon Scattering in the MicroBooNE Detector

We report the first double-differential cross section measurement of neutral-current neutral pion (NC⁢𝜋 0 ) production in neutrino-argon scattering, as well as single-differential measurements of the same channel in terms of final states with and without protons. The kinematic variables of interest for these measurements are the 𝜋 0 momentum and the 𝜋 0 scattering angle with respect to the neutrino beam. A total of 4971 candidate NC⁢𝜋 0 events fully contained within the MicroBooNE detector are selected using data collected at a mean neutrino energy of ∼0.8 GeV from 6.4 ×10 20 protons on target from the Booster Neutrino Beam at the Fermi National Accelerator Laboratory. After extensive data-driven model validation to ensure unbiased unfolding, the Wiener-singular-value-decomposition method is used to extract nominal flux-averaged cross sections. The results are compared to predictions from commonly used neutrino event generators, which tend to overpredict the measured NC⁢𝜋 0 cross section, especially in the 0.2–0.5 GeV/c 𝜋 0 momentum range and at forward scattering angles. Events with at least one proton present in the final state are also underestimated. This data will help improve the modeling of NC⁢𝜋 0 production, which represents a major background in measurements of charge-parity violation in the neutrino sector and in searches for new physics beyond the standard model.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2022

An indirect approach to optimize the reaction rates of thermal NO formation for diesel engines

With stringent emission regulations, it has become more important for modern diesel engine manufacturers to accurately predict engine-out nitrogen oxide (NO x ) emissions across a wide range of operating conditions. Thermal NO is the major source of engine-out NO x in modern diesel engines. For thermal NO formation, several earlier studies have recommended the forward and reverse reaction rate coefficients of the rate-limiting reaction (O + N 2 ⇌ NO + N). However, due to deficiencies in sub-models and inadequacies of reduced chemical mechanisms to represent diesel combustion, these recommended values more often than not need to be adjusted in reduced order combustion models to accurately predict engine-out NO x . Hence, in this work a systematic and computationally efficient approach has been proposed to streamline the process of determining the optimum reaction rate coefficients. Here, to develop the optimization approach, four different production diesel engines with different operating conditions in terms of speed, load, and exhaust-gas recirculation have been considered. Numerical simulations have been performed using a detailed zero-dimensional velocity-composition-frequency transported probability density function (0D-VCF-tPDF) model that uses hundreds of notional particles to capture in-cylinder stratification. Four different combinations of hydrocarbon and NO x chemical mechanisms were used to represent chemistry. It was found that for the rate-limiting reaction, the pre-exponent factors (A f1 , A r1 ) and activation energies (E A,f1 , E A,r1 ) of the forward and reverse reaction rates follow a linear band in A f1 - E A,f1 and A r1 - E A,r1 space where predicted engine-out NO x match the measured values closely. By encompassing such bands from different engines and considering constraints on activation energies, a reduced search domain of pre-exponent factors and activation energies was constructed that is expected to be applicable to any diesel engine. Eventually, computationally efficient three-line and one-line search approaches were proposed to determine the optimum values of the pre-exponent factors and activation energies that led to a minimum error between measured and predicted engine-out NO x . Finally, these three-line and one-line NO x optimization approaches were applied to a fifth production diesel engine for which the 0D-VCF-tPDF model showed a very good predictive performance in terms of predicting peak pressure, 50% burn rate, and engine-out NO x when compared to measured and 3D-CFD values.

33 ADVANCED PROPULSION SYSTEMS↗

DOE OSTI2026

Latent Twins

Over the past decade, scientific machine learning has transformed the development of mathematical and computational frameworks for analyzing, modeling, and predicting complex systems. From inverse problems to numerical partial differential equations (PDEs), dynamical systems, and model reduction, these advances have pushed the boundaries of what can be simulated. Yet they have often progressed in parallel, with representation learning and algorithmic solution methods evolving largely as separate pipelines. With Latent Twins, we propose a unifying mathematical framework that creates a hidden surrogate in latent space for the underlying equations. Whereas digital twins mirror physical systems in the digital world, Latent Twins mirror mathematical systems in a learned latent space governed by operators. Through this lens, classical modeling, inversion, model reduction, and operator approximation all emerge as special cases of a single principle. We establish the fundamental approximation properties of Latent Twins for both ordinary differential equations (ODEs) and PDEs and demonstrate the framework across three representative settings: (i) canonical ODEs, capturing diverse dynamical regimes; (ii) a PDE benchmark using the shallow-water equations, contrasting Latent Twin simulations with deep operator network and forecasts with a four-dimensional variational method baseline; and (iii) a challenging real-data geopotential reanalysis dataset, reconstructing and forecasting from sparse, noisy observations. Latent Twins provide a compact, interpretable surrogate for solution operators that evaluate across arbitrary time gaps in a single-shot, while remaining compatible with scientific pipelines such as assimilation, control, and uncertainty quantification. Looking forward, this framework offers scalable, theory-grounded surrogates that bridge data-driven representation learning and classical scientific modeling across disciplines.

Latent Twins↗

DOE OSTI2023

Comprehensive Review of DPF-based Flash Neutron Radiography Viability

MJOLNIR (MegaJOule Neutron Imaging Radiography) is a Dense Plasma Focus (DPF) being developed by LLNL as a prototype to assess the viability of a DPF as a flash neutron radiography source. To date, MJOLNIR has discharged up to 1.3 MJ of stored energy into a deuterium plasma load and achieved yields up to 1.2 × 10 12 neutrons per discharge. The MJOLNIR pulse length already meets the preliminary requirements for flash neutron radiography and we demonstrate in this report plausible paths forward to meet the yield requirement in a deuterium plasma. MJOLNIR’s neutron spot size has been characterized to be a factor of two greater than radiography requirements and presents the greatest challenge. A path forward to sufficiently shrinking spot size is outlined in this report, using a combination of gas dopants and a smaller anode implosion radius. A DPF is a plasma device with coaxial electrodes whose discharge ends with a stagnated hot and dense plasma column on-axis, at the tip of the central anode. Inside the MJOLNIR DPF plasma column, both thermal and beam target processes generate neutrons. To understand the underlying physics of neutron generation inside the DPF, we model the experiment using a combination of kinetic, fluid, and reduced-order models we have been developing for over ten years at LLNL. Our numerical tools are also pivotal to charting the path forward because they allow us to numerically test out which modifications would improve the DPF’s characteristic output.

42 ENGINEERING↗

DOE OSTI2020

SO(3)-invariance of informed-graph-based deep neural network for anisotropic elastoplastic materials

This work examines the frame-invariance (and the lack thereof) exhibited in simulated anisotropic elasto-plastic responses generated from supervised machine learning of classical multi-layer and informed-graph-based neural networks, and proposes different remedies to fix this drawback. The inherent hierarchical relations among physical quantities and state variables in an elasto-plasticity model are first represented as informed, directed graphs, where three variations of the graph are tested. While feed-forward neural networks are used to train path-independent constitutive relations (e.g., elasticity), recurrent neural networks are used to replicate responses that depends on the deformation history, i.e. or path dependent. In dealing with the objectivity deficiency, we use the spectral form to represent tensors and, subsequently, three metrics, the Euclidean distance between the Euler Angles, the distance from the identity matrix, and geodesic on the unit sphere in Lie algebra, can be employed to constitute objective functions for the supervised machine learning. In this, the aim is to minimize the measured distance between the true and the predicted 3D rotation entities. Following this, we conduct numerical experiments on how these metrics, which are theoretically equivalent, may lead to differences in the efficiency of the supervised machine learning as well as the accuracy and robustness of the resultant models. Neural network models trained with tensors represented in component form for a given Cartesian coordinate system are used as a benchmark. Our numerical tests show that, even given the same amount of information and data, the quality of the anisotropic elasto-plasticity model is highly sensitive to the way tensors are represented and measured. The results reveal that using a loss function based on geodesic on the unit sphere in Lie algebra together with an informed, directed graph yield significantly more accurate rotation prediction than the other tested approaches.

42 ENGINEERING↗

DOE OSTI2024

High-Throughput Screening and Accurate Prediction of Ionic Liquid Viscosities Using Interpretable Machine Learning

Ionic liquids (ILs) are a novel group of green solvents with great promise for various industrial applications, including carbon capture and lignocellulosic biomass deconstruction. However, the use of ILs at the industrial scale remains challenging due to their high viscosities at ambient temperatures. To develop ILs with lower viscosities, a systematic study of their quantitative structure–property relationship (QSPR) is desirable. Here, we developed four machine learning (ML) models to predict viscosity at various temperature and pressure ranges, trained over a wide range of ILs consisting of various cationic and anionic families. ML methods including two-factor polynomial regression (two-factor PR), support vector regression (SVR), feed-forward neural networks (FFNN), and categorical boosting (CATBoost) were developed based on features that have proven useful in previous ML studies: COSMO-RS (conductor-like screening model for real solvents)-derived surface screening charge densities (sigma profiles). FFNN and CATBoost were the most accurate in predicting IL viscosities with lower average absolute relative deviation and higher R2 values on the test set. Tanimoto similarity scores were calculated to characterize the chemical space and structural similarity of the investigated ions. Furthermore, SHapley Additive exPlanation (SHAP) analysis was employed to interpret the ML results. Temperature, the polar area of ILs, and the nonpolar regions of ions are key features that influence the viscosity predictions. Importantly, the IL viscosity prediction here is the most accurate reported to date.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2025

ReSpike: A Co-Design Framework for Evaluating SNNs on ReRAM-Based Neuromorphic Processors

With Moore’s law approaching its end, traditional von Neumann architectures are struggling to keep up with the exceeding performance and memory requirements of artificial intelligence and machine learning algorithms. Unconventional computing approaches such as neuromorphic computing that leverage spiking neural networks (SNNs) to perform computation are gaining traction and seek the paradigm shift necessary to sustain the increasing demands of modern applications. Novel memory technologies, such as resistive RAM (ReRAM), employ a crossbar architecture that possesses the inherent capability of efficiently computing vector-matrix multiplication—a dominant operation in SNNs. The prospect of naturally mapping SNNs to the crossbar structures provides a unique opportunity for achieving a high-performance, power-efficient neuromorphic system. In this work, we present ReSpike, which is a new framework, behavioral simulator, and architectural design based on ReRAM crossbar architectures, enabling modeling and co-design to achieve efficient execution of SNNs. We drive this co-design forward by quantifying the impact that ReRAM cell nonidealities have on the corresponding accuracy of an SNN application.

Asifuzzaman, Kazi [ORNL] (ORCID:0000000240044791)↗