Search NASA⌕ Search

SEARCH · Search NASA

Results for “adaptive weights”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

HydroDCM: Hydrological Domain-Conditioned Modulation for Cross-Reservoir Inflow Prediction

Deep learning models have shown promise in reservoir inflow prediction, yet their performance often deteriorates when applied to different reservoirs due to distributional differences, referred to as the domain shift problem. Domain generalization (DG) solutions aim to address this issue by extracting domain-invariant representations that mitigate errors in unseen domains. However, in hydrological settings, each reservoir exhibits unique inflow patterns, while some metadata beyond observations like spatial information exerts indirect but significant influence. This mismatch limits the applicability of conventional DG techniques to many-domain hydrological systems. To overcome these challenges, we propose HydroDCM, a scalable DG framework for cross-reservoir inflow forecasting. Spatial metadata of reservoirs is used to construct pseudo-domain labels that guide adversarial learning of invariant temporal features. During inference, HydroDCM adapts these features through light-weight conditioning layers informed by the target reservoir’s metadata, reconciling DG’s invariance with location-specific adaptation. Experiment results on 30 real-world reservoirs in the Upper Colorado River Basin demonstrate that our method substantially outperforms state-of-the-art DG baselines under many-domain conditions and remains computationally efficient.

Hu, Pengfei [ORNL] (ORCID:0009000367130950)↗
DOE OSTI2025

Toward particle accelerator machine state embeddings as a modality for large language models

Understanding and diagnosing the state of a particle accelerator requires navigating high-dimensional control system data, often involving hundreds of interdependent parameters. We propose a novel multimodal embedding framework that jointly learns representations of machine states from both numerical control system readouts and natural language descriptions. This enables the translation of complex machine conditions into human-readable summaries while maintaining fidelity to the underlying physical system. The obtained embeddings are subsequently adapted to an open-weights large language model via cross-attention conditioning. We demonstrate a first implementation trained on European XFEL machine state data. This work covers the embedding model architecture, training methodology, and presents initial examples demonstrating the model's capabilities in action. Due to the general concept of machine state, the model can be easily adapted to other facilities and control system environments.

Accelerator Physics↗
DOE OSTI2020

Synchronization-Based Guaranteed-Performance Formation Design for Swarm Systems

Guaranteed-performance formation control for swarm systems with the second-order dynamics is investigated based on the synchronization control strategy. Firstly, a new formation protocol is presented, where the weights of connected edges are adaptively regulated and the performance constraint is imposed. Then, on the basis of the Riccati inequality, sufficient conditions for synchronization-based guaranteed-performance formation are proposed, and an explicit expression of the guaranteed-performance cost is shown, where it is fully distributed to design gain matrices of the formation protocol in the sense that it is independent of global information of swarm systems. Moreover, the whole motion of a swarm system is determined, which is associated with initial states of all agents and formation control vectors. Finally, two numerical examples are shown to demonstrate theoretical conclusions, where the static whole motion and the dynamic whole motion are considered, respectively.

Dang, Hongtao↗
DOE OSTI2025

FM4NPP: A Scaling Foundation Model for Nuclear and Particle Physics

Large language models have revolutionized artificial intelligence by enabling large, generalizable models trained through self-supervision. This paradigm has inspired the development of scientific foundation models (FMs). However, applying this capability to experimental particle physics is challenging due to the sparse, spatially distributed nature of detector data, which differs dramatically from natural language. This work addresses if an FM for particle physics can scale and generalize across diverse tasks. We introduce a new dataset with more than 11 million particle collision events and a suite of downstream tasks and labeled data for evaluation. We propose a novel self-supervised training method for detector data and demonstrate its neural scalability with models that feature up to 188 million parameters. With frozen weights and task-specific adapters, this FM consistently outperforms baseline models across all downstream tasks. The performance also exhibits robust data-efficient adaptation. Further analysis reveals that the representations extracted by the FM are task-agnostic but can be specialized via a single linear mapping for different downstream tasks.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2026

LAF-Net: A Deep Residual and Cross-Attention Framework for Day-Ahead Load Forecasting: Preprint

Accurate day-ahead load forecasting is essential for reliable power system operations and market efficiency. System operators such as the Midcontinent Independent System Operator (MISO) rely on forecasts from multiple vendors, yet combining them effectively remains a persistent challenge due to vendor-specific biases. This paper presents a novel LSTM-Attention Fusion Network with Error Representation (LAF-Net) that enhances day-ahead hourly load forecasting through deep residual learning and multi-modal cross-attention. The proposed model builds a historical error memory from past vendor performance and dynamically queries it with future hour context to generate adaptive, hour-specific trust weights for each vendor. A bounded residual correction further refines forecasts by mitigating systematic and temporally localized errors. Tested on real MISO LBA data with multi-vendor forecasts, LAF-Net consistently outperforms the best vendor baseline across all 38 LBAs, achieving more than a 40% reduction in system-level mean absolute error (MAE) during peak load hours relative to the best vendor baseline.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2021

Data-Driven Forgetting and Discount Factors for Vehicle Speed Forecasting in Ecological Adaptive Cruise Control

Abstract This paper investigates temporal correlations in human driving behavior using real-world driving to improve speed forecasting accuracy. These correlations can point to a measurement weighting function with two parameters: a forgetting factor for past speed measurements that the vehicle itself drove with, and a discount factor for the speeds of vehicles ahead based on information from vehicle-to-vehicle communication. The developed weighting approach is applied to a vehicle speed predictor using polynomial regression, a prediction method well-known in the literature. The performance of the developed approach is then assessed in both real-world and simulated traffic scenarios for accuracy and robustness. The new weighting method is applied to an ecological adaptive cruise control system, and its influence is analyzed on the prediction accuracy and the performance of the ecological adaptive cruise control in an electric vehicle powertrain model. The results show that the new prediction method improves energy saving from the eco-driving by up to 4.7% compared to a baseline least-square-based polynomial regression. This is a 10% improvement over the constant speed/acceleration model, a conventional speed predictor.

Automation & Control Systems↗
DOE OSTI2023

High-Order Mesh Morphing for Boundary and Interface Fitting to Implicit Geometries

Here, we propose a method that morphs high-order meshes such that their boundaries and interfaces coincide/align with implicitly defined geometries. Our focus is particularly on the case when the target surface is prescribed as the zero isocontour of a smooth discrete function. Common examples of this scenario include using level set functions to represent material interfaces in multimaterial configurations, and evolving geometries in shape and topology optimization. The proposed method formulates the mesh optimization problem as a variational minimization of the sum of a chosen mesh-quality metric using the Target-Matrix Optimization Paradigm (TMOP) and a penalty term that weakly forces the selected faces of the mesh to align with the target surface. The distinct features of the method are use of a source mesh to represent the level set function with sufficient accuracy, and adaptive strategies for setting the penalization weight and selecting the faces of the mesh to be fit to the target isocontour of the level set field. We demonstrate that the proposed method is robust for generating boundary- and interface-fitted meshes for curvilinear domains using different element types in 2D and 3D.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Efficient Multiscale Optoelectronic Prediction for Conjugated Polymers

Conjugated polymers represent a high-potential material class due to the tunability of their optoelectronic properties via straightforward processing protocols. However, to in silico tailor these optoelectronic properties, one must compute the conformationally dependent electronic structure over mesoscopic length and time scales. This task represents a challenging multiscale computational problem in which quantum chemistry and atomistic or coarse-grained molecular dynamics must be integrated. Recently, we introduced a methodology, called electronic coarse graining (ECG), that utilizes artificial neural networks to compute ab initio quality electronic structures using only the reduced degrees of freedom of coarse-grained models. Here, we adapt ECG to variable-molecular weight conjugated oligomers by casting the problem in the framework of sequence detection. A machine-learning (ML) architecture employing one-dimensional convolutional neural networks and long short-term memory networks is utilized to compute ground-state orbital energies, charge density distributions, and optical spectra solely from the coarse-grained model's configurational degrees of freedom. Robust molecular weight transferability for ECG is established via a A-ML approach that leverages model electronic Hamiltonians for ground and excited-state property determination. The accuracy and transferability of the ECG methodology opens the door for scalable optoelectronic property prediction in conjugated polymers directly from coarse-grained degrees of freedom.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Scaling Out a Combinatorial Algorithm for Discovering Carcinogenic Gene Combinations to Thousands of GPUs

Cancer is a leading cause of death in the US, second only to heart disease. It is primarily a result of a combination of an estimated two-nine genetic mutations (multi-hit combinations). Although a body of research has identified hundreds of cancer-causing genetic mutations, we don’t know the specific combination of mutations responsible for specific instances of cancer for most cancer types. An approximate algorithm for solving the weighted set cover problem was previously adapted to identify combinations of genes with mutations that may be responsible for individual instances of cancer. However, the algorithm’s computational requirement scales exponentially with the number of genes, making it impractical for identifying more than three-hit combinations, even after the algorithm was parallelized and scaled up to a V100 GPU. Since most cancers have been estimated to require more than three hits, we scaled out the algorithm to identify combinations of four or more hits using 1000 nodes (6000 V100 GPUs with ≈48×106 processing cores) on the Summit supercomputer at Oak Ridge National Laboratory. Efficiently scaling out the algorithm required a series of algorithmic innovations and optimizations for balancing an exponentially divergent workload across processors and for minimizing memory latency and inter-node communication. We achieved an average strong scaling efficiency of 90.14% (80.96%–97.96% for 200 to 1000 nodes), compared to a 100 node run, with 84.18% scaling efficiency for 1000 nodes. With experimental validation, the multi-hit combinations identified here could provide further insight into the etiology of different cancer subtypes and provide a rational basis for targeted combination therapy.

Dash, Sajal↗
DOE OSTI2025

Wellbore Stability and Mud Loss Management in Geothermal Drilling: Optimizing Mud Weight to Mitigate Tensile Wellbore Fracturing at The Geysers, California

As part of a U.S. Department of Energy (DOE) Geothermal Technologies Office-funded initiative, Geysers Power Company, LLC, a subsidiary of Calpine Corporation, has been working to enhance drilling performance at the world’s largest geothermal field, The Geysers, in northern California. In a recent drilling operation of the GDC-36 well, excessive mud losses were encountered, initially addressed through repeated but largely ineffective cement plugging. Ultimately, the most effective strategy was to drill blind through the loss zones, made feasible by the high rate of penetration (ROP) achieved with PDC bits, allowing significant progress before the mud tanks were depleted and water-sensitive argillic formation layers could collapse. In response to these challenges, the project team explored alternative methods to minimize downtime and risks associated with cement plugging and continuous mud loss and to contemplate the driving mechanisms for the losses. Wellbore imaging using Formation MicroImager (FMI) and Ultrasonic Borehole Imager (UBI) tools revealed longitudinal tensile fractures, which were attributed to mud weights exceeding the minimum circumferential stress resulting from the native stress field and formation pressure. This study examines the mud losses encountered and leverages wellbore imaging data to understand the mechanisms behind mud induced tensile fracturing in specific rock facies. Understanding fracture behavior across different lithologies is crucial, as fractures within the reservoir can enhance steam migration throughout the system. The reservoir at The Geysers lies within the Mesozoic Franciscan Assemblage, a tectonic mélange formed by subduction. It consists of metamorphosed turbidite sandstone (greywacke) and mudstone (argillite), oceanic upper crust (including greenstone and chert), and serpentinized ultramafic rocks - each exhibiting distinct geomechanical fracturing properties. The structural fabric of the Franciscan Assemblage was shaped by low-angle Mesozoic thrust faulting and later overprinted by sub-vertical strike-slip structures related to the Pacific-North American plate boundary. A wellbore stability model was developed using core measurements and logs to simulate fracturing scenarios during drilling under varying stress conditions. These simulations guided the development of an optimized mud weight management strategy that should enable adaptive adjustments during drilling, reducing the likelihood of tensile fracturing and mud losses, ultimately improving operational efficiency.

15 GEOTHERMAL ENERGY↗
DOE OSTI2022

The contribution of the neuroendocrine system to adaption after repeated daily ozone exposure in rats

Ozone-induced lung injury/inflammation dissipates despite continued exposure for 3 or more days; however, the mechanisms of adaptation/habituation remain unclear. Since ozone effects are mediated through adrenal-derived stress hormones, which also regulate longevity of centrally-mediated stress response, we hypothesized that ozone-adaptation is linked to diminution of neuroendocrine stress-axes activation and glucocorticoid levels. Male Wistar-Kyoto-rats (12-week-old) were injected with vehicle or a therapeutically-relevant dexamethasone dose (0.01-mg/kg/day; intraperitoneal) for 1-month to determine if suppression of glucocorticoid signaling was linked to adaptation. Vehicle- and dexamethasone-treated rats were exposed to air or 0.8-ppm ozone, 4 h/day × 2 or 4 days to assess the impacts of acute exposure and adaptation, respectively. Dexamethasone reduced thymus and spleen weights, circulating lymphocytes, corticosterone and increased insulin. Ozone increased lavage-fluid protein and neutrophils and decreased circulating lymphocytes at day-2 but not day-4. Ozone-induced hyperglycemia, glucose intolerance and inhibition of beta-cell insulin release occurred at day-1 but not day-3. Ozone depleted circulating prolactin, thyroid-stimulating hormone, and luteinizing-hormone at day-2 but not day-4, suggesting central mediation of adaptation. Adrenal epinephrine biosynthesis gene, Pnmt, was up-regulated after ozone exposure at both timepoints. However, genes involved in glucocorticoid biosynthesis were up-regulated after day-2 but not day-4, suggesting that acute 1- or 2-day ozone-mediated glucocorticoid increase elicits feedback inhibition to dampen hypothalamic stimulation of ACTH release in response to repeated subsequent ozone exposures. Although dexamethasone pretreatment affected circulating insulin, lymphocytes and adrenal genes, it had modest effect on ozone adaptation. In conclusion, ozone adaptation likely involves lack of hypothalamic response due to reduced availability of circulating glucocorticoids.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2020

Stochasticity and robustness in spiking neural networks

Despite drawing inspiration from biological systems which are inherently noisy and variable, artificial neural networks have been shown to require precise weights to carry out the task which they are trained to accomplish. This creates a challenge when adapting these artificial networks to specialized execution platforms which may encode weights in a manner which restricts their accuracy and/or precision.Reflecting back on the non-idealities which are observed in biological systems, we investigated the effect these properties have on the robustness of spiking neural networks under perturbations to weights. First, we examined techniques extant in conventional neural networks which resemble noisy processes, and postulated they may produce similar beneficial effects in spiking neural networks. Second, we evolved a set of spiking neural networks utilizing biological non-idealities to solve a pole-balancing task, and estimated their robustness. We showed it is higher in networks using noisy neurons, and demonstrated that one of these networks can perform well under the variance expected when a hafnium-oxide based resistive memory is used to encode synaptic weights. Lastly, we trained a series of networks using a surrogate gradient method on the MNIST classification task. We confirmed that these networks demonstrate similar trends in robustness to the evolved networks. We discuss these results and argue that they display empirical evidence supporting the role of noise as a regularizer which can increase network robustness.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Adaptive spectra-to-exposure conversion using ridge regularized polynomial response models

Real-time gamma spectra-to-exposure conversion in aerial and ground monitoring commonly relies on calibration-derived, detector- or system-specific conversion coefficients that are assumed to generalize across operational environments. In practice, deployment specific differences in spectral composition and transport conditions can introduce systematic bias relative to reference instruments, motivating methods that adapt coefficients using minimal field supervision while explicitly limiting overfitting. In this work, we present a conservative coefficient adaptation framework that updates a baseline polynomial energy-weighting function using ridge-regularized regression, with leave-one-out cross-validation (LOOCV) used to select the regularization strength. The findings support ridge-constrained minimal-supervision adaptation as a practical mechanism to suppress site-specific bias without destabilizing a calibration-derived baseline.

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2022

A Decision-Making Machine Learning Approach in Hermite Spectral Approximations of Partial Differential Equations

The accuracy and effectiveness of Hermite spectral methods for the numerical discretization of partial differential equations on unbounded domains are strongly affected by the amplitude of the Gaussian weight function employed to describe the approximation space. This is particularly true if the problem is under-resolved, i.e., there are no enough degrees of freedom. The issue becomes even more crucial when the equation under study is time-dependent, forcing in this way the choice of Hermite functions where the corresponding weight depends on time. In order to adapt dynamically the approximation space, it is here proposed an automatic decision-making process that relies on machine learning techniques, such as deep neural networks and support vector machines. The algorithm is numerically tested with success on a simple 1D problem, but the main goal is its exportability in the context of more serious applications. Here we also show at the end an application in the framework of plasma physics.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

A robust alloy design (RAD) strategy for next-generation (IV) nuclear fission reactors

Next-generation nuclear reactors demand structural materials capable of withstanding extreme conditions, including high temperatures, intense neutron flux, and corrosive environments. Multi-Principal Element Alloys (MPEAs) have emerged as promising candidates due to their exceptional radiation tolerance, thermal stability, and compositional flexibility. This study introduces a versatile and customizable Robust Alloy Design (RAD) strategy for systematically designing MPEAs for GEN-IV reactor fuel cladding. The RAD framework integrates nuclear-relevant selection criteria, empirical parameter assessments, and high-throughput CALPHAD simulations to efficiently narrow compositional space and identify stable alloys. A unified RAD score developed for the first time, combines key performance metrics, including fuel-clad chemical interaction (FCCI), neutron absorption cross-section (NAC), valence electron configuration (VEC), and melting point factor (MPF), into a flexible ranking system adaptable to reactor-specific priorities. Among 724 candidates, V555(5Al–5Cr–5Fe–85V) emerged as the top alloy, validated experimentally with a homogeneous single-phase BCC microstructure and superior mechanical properties (nano-indentation: 3.389 ± 0.258 GPa; Vickers hardness: 240 ± 6.7 HV), significantly outperforming Zircaloy-4 and V-4Cr-4Ti. Importantly, the RAD strategy is not limited to nuclear applications; its customizable weighting system enables scalability to other extreme environments. This adaptability positions RAD strategy as a versatile tool for advanced materials design across multiple industries.

Alloy design↗
DOE OSTI2023

TransformerG2G: Adaptive time-stepping for learning temporal graph embeddings using transformers

Dynamic graph embedding has emerged as a very effective technique for addressing diverse temporal graph analytic tasks (i.e., link prediction, node classification, recommender systems, anomaly detection, and graph generation) in various applications. Such temporal graphs exhibit heterogeneous transient dynamics, varying time intervals, and highly evolving node features throughout their evolution. Hence, incorporating long-range dependencies from the historical graph context plays a crucial role in accurately learning their temporal dynamics. In this paper, we develop a graph embedding model with uncertainty quantification, TransformerG2G, by exploiting the advanced transformer encoder to first learn intermediate node representations from its current state (t) and previous context (over timestamps [t–1,t–l], l is the length of context). Moreover, we employ two projection layers to generate lower-dimensional multivariate Gaussian distributions as each node's latent embedding at timestamp t. We consider diverse benchmarks with varying levels of "novelty" as measured by the TEA (Temporal Edge Appearance) plots. Here, our experiments demonstrate that the proposed TransformerG2G model outperforms conventional multi-step methods and our prior work (DynG2G) in terms of both link prediction accuracy and computational efficiency, especially for high degree of novelty. Furthermore, the learned time-dependent attention weights across multiple graph snapshots reveal the development of an automatic adaptive time stepping enabled by the transformer. Importantly, by examining the attention weights, we can uncover temporal dependencies, identify influential elements, and gain insights into the complex interactions within the graph structure. For example, we identified a strong correlation between attention weights and node degree at the various stages of the graph topology evolution.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Identification of Differential Equations by Dynamics-Guided Weighted Weak Form with Voting

In the identification of differential equations from data, significant progresses have been made with the weak/integral formulation. In this paper, we explore the direction of finding more efficient and robust test functions adaptively given the observed data. While this is a difficult task, we propose weighting a collection of localized test functions for better identification of differential equations from a single trajectory of noisy observations on the differential equation. We find that using high dynamic regions is effective in finding the equation as well as the coefficients, and propose a dynamics indicator per differential term and weight the weak form accordingly. For stable identification against noise, we further introduce a voting strategy to identify the active features from an ensemble of recovered results by selecting the features that frequently occur in different weighting of test functions. Systematic numerical experiments are provided to demonstrate the robustness of our method.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

SA-GAT-SR: self-adaptable graph attention networks with symbolic regression for high-fidelity material property prediction

Recent advances in machine learning have demonstrated an enormous utility of deep learning approaches, particularly Graph Neural Networks (GNNs) for materials science. These methods have emerged as powerful tools for high-throughput prediction of material properties, offering a compelling enhancement and alternative to traditional first-principles calculations. While the community has predominantly focused on developing increasingly complex and universal models to enhance predictive accuracy, such approaches often lack physical interpretability and insights into materials behavior. Here, we introduce a novel computational paradigm—Self-Adaptable Graph Attention Networks integrated with Symbolic Regression (SA-GAT-SR)—that synergistically combines the predictive capability of GNNs with the interpretative power of symbolic regression. Our framework employs a self-adaptable encoding algorithm that automatically identifies and adjust attention weights so as to screen critical features from an expansive 180-dimensional feature space while maintaining O(n) computational scaling. The integrated SR module subsequently distills these features into compact analytical expressions that explicitly reveal quantum-mechanically meaningful relationships, achieving 23 × acceleration compared to conventional SR implementations that heavily rely on first-principle calculations-derived features as input. This work suggests a new framework in computational materials science, bridging the gap between predictive accuracy and physical interpretability, offering valuable physical insights into material behavior.

36 MATERIALS SCIENCE↗