SEARCH · Engineering Papers
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.
This poster details the evaluation of a Michel electron selection algorithm centered on the neural network-based software SPINE (Scalable Particle Imageing with Neural Embeddings). The algorithm was developed and calibrated using simulated data from the SBND (Short-Baseline Neutrino Detector) experiment before being applied to simulated data from DUNE (Deep Underground Neutrino Experiment).
Michel electrons are a valuable input for particle detector calibration due to their consistent kinetic energy distribution. This report details the evaluation of a Michel electron identification method's application to simulated data from the DUNE (Deep Underground Neutrino Experiment) far detector. This method, which relies on the neural network-based particle classification software SPINE (Scalable Particle Imaging with Neural Embeddings), was developed and calibrated using simulated data for the SBND (Short-Baseline Neutrino Detector) experiment before being applied to simulated DUNE data from a 1x2x6 subset of far detector modules.
The complex events observed at the NOvA long-baseline neutrino oscillation experiment contain vital information for understanding the most elusive particles in the standard model. The NOvA detectors observe interactions of neutrinos from the NuMI beam at Fermilab. Associating the particles produced in these interaction events to their source particles, a process known as reconstruction, is critical for accurately measuring key parameters of the standard model. Events may contain several particles, each producing sparse high-dimensional spatial observations, and current methods are limited to evaluating individual particles. To accurately label these numerous, high-dimensional observations, we present a novel neural network architecture that combines the spatial learning enabled by convolutions with the contextual learning enabled by attention. This joint approach, TransformerCVN, simultaneously classifies each event and reconstructs every individual particle's identity. TransformerCVN classifies events with 90% accuracy and improves the reconstruction of individual particles by 6% over baseline methods which lack the integrated architecture of TransformerCVN. In addition, this architecture enables us to perform several interpretability studies which provide insights into the network's predictions and show that TransformerCVN discovers several fundamental principles that stem from the standard model.
Machine learning (ML) methods, especially reinforcement learning (RL), have been widely considered for traffic signal optimization in intelligent transportation systems. Most of these ML methods are centralized, lacking in scalability and adaptability in large traffic networks. Further, it is challenging to train such ML models due to the lack of training platforms and/or the cost of deploying and training in a real traffic networks. This paper presents an approach for the integration of decentralized graph-based multi-agent reinforcement learning (DGMARL) with a Digital Twin (DT) to optimize traffic signals for the reduction of traffic congestion and network-wide fuel consumption related to stopping. Specifically, the DGMARL agents learn traffic state patterns and make decisions regarding traffic signal control with assistance from a Digital Twin module, which simulates and replicates the traffic behaviors of a real traffic network. The proposed approach was evaluated using PTV-Vissim [1], a microscopic traffic simulation platform. PTV-Vissim is also the simulation engine of the DT, enabling emulation and optimization of the traffic signals on the MLK Smart Corridor in Chattanooga, Tennessee. Compared to an actuated signal control baseline approach, experiment results show that Eco_PI, a developed performance measure capturing the impact of stops on fuel consumption, was reduced by 44.27% in a 24-hour and an average of 29.88% in a PM-peak-hour scenario.
Detecting anomalous sequences is an integral part of building and protecting modern large-scale health information technology (HIT) systems. These HIT systems generate a large volume of records of patients’ state and significant events, which provide a valuable resource to help improve clinical decisions, patient care processes, and other issues. However, detecting anomalous sequences in electronic health records (EHR) remains a challenge in healthcare applications for several reasons, including imbalances in the data, complexity of relationships between events in the sequence, and the curse of dimensionality. Conventional anomaly detection methods use the finite sequence of events to discriminate sequences. They fail to incorporate salient event details under variable higher-order dependencies (e.g., duration between events) that can provide better discrimination of sequences in their models. To address this problem, we propose event sequence and subsequence anomaly detection algorithms that (1) use network-based representations of interactions in the data, (2) account for variable higher-order dependencies in the data, and (3) incorporate events duration for adequate discrimination of the data. The proposed approach identifies anomalies by monitoring the change in the graph after the test sequence is removed from the network. The change is quantified using graph distance metrics so that dramatic changes in the network can be attributed to the removed sequence. Furthermore, the proposed subsequence algorithm recommends plausible paths and salient information for the detected anomalous subsequences. Our results show that the proposed event sequence anomaly detection algorithm outperforms the baseline methods for both synthetic data and real-world EHR data.
The Cray HPE Slingshot 11 network is used on the new exascale systems arriving at the U.S. Department of Energy (DoE) laboratories (e.g., Frontier, Aurora, Perlmutter). As such, the support of this network is an important capability to meet the needs of exascale applications. Here, this article highlights recent work to develop supporting infrastructure to enable Open MPI to efficiently support these new platforms. A key component of this effort involves development of a new Open Fabrics Interface (OFI) provider, LinkX. We discuss the design and development of enhancements that take advantage of the new Slingshot 11 network and AMD GPUs. We include performance data from tests on the Frontier supercomputer using synthetic communication benchmarks, and the vendor provided MPI as a baseline for comparison. The tests demonstrate full functionality of Open MPI on the system and initial results show favorable performance when compared to the highly tuned vendor implementation.
Code repository for the experiments performed in the paper: On the Effectiveness of Neural Operators at Zero-Shot Weather Downscaling (https://doi.org/10.1017/eds.2025.11) Overall, our work investigates the zero-shot downscaling potential of neural operators. To summarize, our contributions are: 1. We provide a comparative analysis based on two challenging weather downscaling problems, between various neural operator and non-neural-operator methods with large upsampling factors (e.g., 8x and 15x) and fine grid resolutions (e.g., 2 km × 2 km wind speed). 2. We examine whether neural operator layers provide unique advantages when testing downscaling models on upsampling factors higher than those seen during training, i.e., zero-shot downscaling. Our results instead show the surprising success of an approach that combines a powerful transformer-based model with a parameter-free interpolation step at zero-shot weather downscaling. 3. We find that this Swin-Transformer-based approach mostly outperforms all neural operator models in terms of average error metrics, whereas an enhanced super-resolution generative adversarial network (ESRGAN)-based approach is better than most models in capturing the physics of the system, and suggests their use in future work as strong baselines. However, these approaches still do not capture variations at smaller spatial scales well, including the physical characteristics of turbulence in the HR data. This suggests a potential for improvement in transformer or GAN-based methods and neural-operator-based methods for zero-shot weather downscaling.
The recently released Circularity Gap Report 2023 by Circle Economy Foundation found that the circularity score for the global economy is declining. This indicates that material extraction from virgin sources is climbing over earlier years. As per data released by US Geological Survey, material use in the US economy has increased exponentially with the expanding US economy over last century. EPA tracked municipal solid waste from 1960 to 2018 and found that 50% of the waste is destined for landfills. EPA estimates that US industry is responsible for 2.7 billion ton of solid non-hazardous waste annually in our mostly linear economy model. The circular economy framework aims at decoupling economic value generation from virgin materials extraction from nature. The linear model of material extraction and disposal at end of life is highly unsustainable. Manufacturing companies have realized this and in their commitments to sustainability are adopting ambitious waste reduction targets. Through Better Plants program US Department of Energy has established Waste Reduction Network where it is offering technical assistance to partners to achieve these ambitious waste reduction goals. Basic requirements of establishing a target include identifying a baseline, quantifying waste performance, and measuring progress over time. One problem faced by industry is the unstandardized metrics to quantify waste performance that may not be well suited to demonstrate progress. In this paper, we research methods traditionally used to measure waste performance and highlight advantages, gaps and limitations of each method. Suitability of the methods applicable to different manufacturing circumstances are also examined. Finally, we present a case study of a large manufacturer that faced inconsistencies in their tracked measurement metric. A solution was proposed to alter the methodology to enable more accurate waste performance tracking against a baseline.
People’s opinions on a wide range of topics often evolve over time through their interactions with others. Models of opinion dynamics primarily focus on one-dimensional opinions, which represent opinions on one topic. However, opinions on various topics are rarely isolated; instead, they can be interdependent and correlated. In a bounded-confidence model (BCM) of opinion dynamics, agents are receptive to each other only if their opinions are sufficiently similar. Here, we extend classical agent-based BCMs—namely, the Hegselmann–Krause BCM, which has synchronous interactions, and the Deffuant–Weisbuch BCM, which has asynchronous interactions—to a multidimensional setting, in which the opinions are multidimensional vectors representing opinions of different topics and opinions on different topics are interdependent. To measure opinion differences between agents, we introduce topic-weighted discordance functions that account for opinion differences in all topics. We define regions of receptiveness for our models, and we use them to characterize the steady-state opinion clusters and provide an analytical approach to compute these regions. In addition, we numerically simulate our models on various networks with initial opinions drawn from a variety of distributions. When initial opinions are correlated across different topics, our topic-weighted BCMs yield significantly different results in both transient and steady states compared to baseline models, where the dynamics of each opinion topic are independent.
The internet is turning out to be an integral part of every-one's lives as more and more devices are being connected to serve societal needs. Our work is motivated by two ma-jor observations. Firstly, one drawback of connecting to the network is the threat of network attacks that can compromise users' private information, leading to data loss and adversely affecting productivity. There are several traditional security mechanisms to defend against these attacks, such as firewalls, virtual private networks (VPNs), demilitarized zones (DMZs), and vulnerability scanners. One way to prevent these attacks is early detection and prevention. However, these kinds of architecture do not scale very well because of their centralized nature. Secondly, we observe from heuristics and data set distributions that the majority of the requests made to a server are innocuous. Therefore, almost all server request data sets are highly imbalanced, weighted highly towards the harmless requests.
Much success was achieved throughout the course of this project. A successful implementation of Dynamic Neural Network Optimization (D-NNO) was coupled with Adaptive Predictive Controls (APC) and a novel hardware installation comprised of an advanced sensor network (ASN) measuring mass-weighted averages of flue gas constituents above the horizontal superheater of a coal-fired utility boiler. From 2019 through 2023 (including an extension due to COVID delays), the team was able to prototype, evaluate, deploy, iterate, and ultimately finalize an advanced closed-loop control D-NNO system which demonstrated the ability to: •improve unit efficiency ~2.0% relative to unoptimized operation (represented as total fuel fired per MWh generated) •improve unit NOx emission rates 10%+ beyond static optimization baselines •improve unit temperature stability as much as 58% and on average 12% •improve operating load stability as much as 35% The culmination of this project has generated an advanced methodology of deploying specially designed recurrent neural networks (long short-term memory, gated recurrent unit, encoder-decoder networks, transformers, etc.), customized trajectory planning and closed-loop optimization modules capable of adapting to live electric grid responses and demands, self-tuning and adaptive expert controls constantly adjusting prediction parameters to real-time unit behavior, and a hardware/software package able to reliably calculate net unit heat rate (NUHR) in real-time using flue gas constituents, machine learning, and known combustion relationships. Through this real-time NUHR value, immediate feedback on system adjustments relative to operating efficiency was available, allowing for rapid improvements to system performance. In addition to development and deployment of the advanced D-NNO system, the approach methodology has been readily commercialized through the project platform Griffin Open Systems, LLC, the D-NNO software platform host. Similar methodologies to those developed by this project have already been deployed at 5 other units across the United States, with another 6 implementations scheduled, and more expected. Over the course of the project, multiple academic papers were submitted and accepted for publication within esteemed academic journals, and PhD students were trained and graduated, as well as undergraduate students becoming involved and participating to project objectives.
In this paper, we explore the application of convolutional neural networks (CNNs) for predicting the chemical composition of complex geologic samples in a simulated Martian atmospheric environment. Specifically, we aim to characterize oxide weight percentages (wt.%) of rock samples analyzed by remote Laser-Induced Breakdown Spectroscopy (LIBS), framing the problem as a multi-target regression task . Neural networks trained on LIBS spectra are prone to overfitting due to high spectral complexity, limited labeled data, and measurement noise. While regularization is critical for improving generalization, common methods (e.g., ℓ 2 regularization) impose constraints not directly tied to data distribution properties. We propose a novel regularization method based on a specific ƒ-divergence induced by a graph-based estimator, designed to constrain the distributional discrepancy between predictions and targets. This regularizer serves a dual purpose: (a) mitigating overfitting by enforcing a constraint on the distributional difference between predictions and noisy targets, and (b) acting as an auxiliary loss that penalizes large divergences. To enable backpropagation, we develop a differentiable approximation of this particular ƒ-divergence, making the method feasible for neural networks. Experiments on ChemCam and SuperCam LIBS calibration spectra show that mathematical equation-divergence regularization outperforms or matches standard regularization methods (ℓ 1 , ℓ 2 , dropout) and the classical baseline, partial least squares (PLS). Combining ƒ-divergence regularization with standard regularization yields further performance gains, indicating that distributional regularization is useful in this context giving a promising direction for robust model training in planetary science applications. Source code is publicly available at Klein and Li (2025), https://doi.org/10.11578/dc.20250530.7.
We introduce and investigate, for finite groups G, G-invariant deep neural network (GDNN) architectures with ReLU activation that are densely connected- i.e., include all possible skip connections. In contrast to other G-invariant architectures in the literature, the preactivations of theG-DNNs presented here are able to transform by signed permutation representations (signed perm-reps) of G. Moreover, the individual layers of the G-DNNs are not required to be G-equivariant; instead, the preactivations are constrained to be G-equivariant functions of the network input in a way that couples weights across all layers. The result is a richer family of G-invariant architectures never seen previously. We derive an efficient implementation of G-DNNs after a reparameterization of weights, as well as necessary and sufficient conditions for an architecture to be "admissible"- i.e., nondegenerate and inequivalent to smaller architectures. We include code that allows a user to build a G-DNN interactively layer-by-layer, with the final architecture guaranteed to be admissible. We show that there are far more admissible G-DNN architectures than those accessible with the "concatenated ReLU" activation function from the literature. Finally, we apply G-DNNs to two example problems--(1) multiplication in --1, 1} (with theoretical guarantees) and (2) 3D object classification--finding that the inclusion of signed perm-reps significantly boosts predictive performance compared to baselines with only ordinary (i.e., unsigned) perm-reps.
Cataloging and classifying trees in the urban environment is a crucial step in urban and environmental planning; however, manual collection and maintenance of this data is expensive and time-consuming. Although algorithmic approaches that rely on remote sensing data have been developed for tree detection in forests, they generally struggle in the more varied urban environment. This work proposes a novel end-to-end deep learning method for the detection of trees in the urban environment from remote sensing data. Specifically, we develop and train a novel PointNet-based neural network architecture to predict tree locations directly from LiDAR data augmented with multi-spectral imagery. We compare this model to a number of high-performing baselines on a large and varied dataset in the Southern California region, and find that our method outperforms all baselines in terms of tree detection ability (75.5% F-score) and positional accuracy (2.28 meter root mean squared error), while being highly efficient. We then analyze and compare the sources of errors, and how these reveal the strengths and weaknesses of each approach. Our results highlight the importance of fusing spectral and structural information for remote sensing tasks in complex urban environments.
FPGA-based SmartNICs offer great potential to significantly improve the performance of high-performance computing and warehouse data processing by tightly coupling support for reconfigurable data-intensive computation with cross-node communication, thereby mitigating the von Neumann bottleneck. Existing work, however, has been generally been limited in that it assumes an accelerator model where kernels are offloaded to SmartNICs, but most control tasks are left to the CPUs. This leads to frequent waiting, inferior performance, and scaling challenges. In this work, we propose a new distributive data-centric computing framework, named FCsN, for reconfigurable SmartNIC-based systems. Through a lightweight task circulation execution model and its implementation architecture, FCsN allows the complete detaching of kernel execution, control logic, system scheduling, and network communication to the SmarNICs. This boosts performance by: (i) avoiding the control dependency with CPUs and (ii) supporting streaming kernel execution and network communication at line rate and in a very fine-grained manner. We demonstrate the efficiency and flexibility of FCsN using various types of neural network applications including graph neural networks; as these last are both irregular and data intensive they offer an especially robust demonstration. Evaluations using commonly-used neural network models and graph datasets show that a system with the support of FCsN can achieve, on average, 144 speedups over the MPI-based standard CPU baselines.
This data package consists of scripts and data that were used to generate all the results and figures for Mital et al., 2020 (https://doi.org/10.3389/frwa.2020.00020). The purpose of the study was to develop a new algorithm to impute (or gap-fill) missing daily precipitation data. Study area was the Upper Colorado Water Resources Region (UCWRR), over a time period of 2008-2017. In terms of data, the package consists of raw precipitation measurements from Natural Resources Conservation Service (NRCS) network. These raw measurements are compiled into a single csv file (NRCS_dates_2008_2017.csv). The corresponding metadata are compiled in Metadata_processed.csv. The package also consists of outputs generated during Baseline and Sequential imputation runs as documented in Mital et al., 2020. Scripts used to generate all the results and figures for Mital et al., 2020 are also uploaded. A readme file documenting the layout of the data archive is also uploaded, and a KML file shows the geographic range for the data.
This report is intended to detail the findings of our investigation of the applicability of machine learning to the task of aftershock identification. The ability to automatically identify nuisance aftershock events to reduce analyst workload when searching for events of interest is an important step in improving nuclear monitoring capabilities and while waveform cross - correlation methods have proven successful, they have limitations (e.g., difficulties with spike artifacts, multiple aftershocks in the same window) that machine learning may be able to overcome. Here we apply a Paired Neural Network (PNN) to a dataset consisting of real, high quality signals added to real seismic noises in order to work with controlled, labeled data and establish a baseline of the PNN's capability to identify aftershocks. We compare to waveform cross - correlation and find that the PNN performs well, outperforming waveform cross - correlation when classifying similar waveform pairs, i.e., aftershocks.
The accelerator-driven beam uncertainty limits oscillation measurements in long-baseline neutrino experiments. Spill-resolved beam diagnostics and real-time inference are necessary to address these neutrino flux systematics. As such, we present a machine-learning-based beam monitoring framework developed and validated using data from the T2K experiment. Our approach uses downstream, spill-by-spill muon monitor observables to predict upstream parameters such as proton beam position and width. We achieve high predictive accuracy on nominal runs, demonstrating robust baseline performance whether the model is trained on stable runs or systematically varied conditions. The framework is designed to be robust against domain shifts, allowing the neural network architectures and inference strategies developed with T2K data to be retrained and validated using LBNF simulations, with the goal of eventual deployment under real LBNF/DUNE operating conditions. This scalable approach to real-time beam inference offers a pathway toward reducing flux systematics for next-generation neutrino experiments such as DUNE.