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.
A general algorithm for determining diabatic representations from adiabatic energies, energy gradients and derivative couplings using neural networks is introduced.
A Python-TensorFlow tool that learns chemical kinetics (LearnCK) for systematic and automatic reduction of microkinetic models using artificial neural networks (NNs).
This review explores the intersection of bio-plausible artificial intelligence in the form of spiking neural networks (SNNs) with the analog in-memory computing (IMC) domain, highlighting their collective potential for low-power edge computing environments. Through detailed investigation at the device, circuit, and system levels, we highlight the pivotal synergies between SNNs and IMC architectures. Additionally, we emphasize the critical need for comprehensive system-level analyses, considering the inter-dependencies among algorithms, devices, circuit, and system parameters, crucial for optimal performance. An in-depth analysis leads to the identification of key system-level bottlenecks arising from device limitations, which can be addressed using SNN-specific algorithm–hardware co-design techniques. This review underscores the imperative for holistic device to system design-space co-exploration, highlighting the critical aspects of hardware and algorithm research endeavors for low-power neuromorphic solutions.
In this work, we present a variational Monte Carlo method that solves the nuclear many-body problem in the occupation number formalism exploiting an artificial neural network representation of the groundstate wave function. A memory-efficient version of the stochastic reconfiguration algorithm is developed to train the network by minimizing the expectation value of the Hamiltonian. We benchmark this approach against widely used nuclear many-body methods by solving a model used to describe pairing in nuclei for different types of interaction and different values of the interaction strength. Despite its polynomial computational cost, our method outperforms coupled-cluster and provides energies that are in excellent agreement with the numerically-exact full configuration interaction values.
Analysis of electrical noise in vanadium oxide electrochemical random access memory devices and its potential for application in a Bayesian neural network
This paper explores the detection and localization of cyber-attacks on power systems, focusing on comparing conventional machine learning (ML) and deep learning methods, and graph neural network (GNN)-based techniques. We assess the detection accuracy of these approaches and their potential to pinpoint the locations of specific buses under attack. Given the demonstrated success of GNNs in other time series anomaly detection applications, we aim to evaluate their performance within the context of power systems cyber-attack. Utilizing the IEEE 68-bus system, we simulated four types of attacks to test the selected approaches. Our results indicate that GNN-based methods outperform conventional machine learning and deep learning models in detection. Additionally, GNNs show promise in accurately localizing attacks for simple scenarios, although they still face challenges in more complex cases.
Momentum is crucial in stochastic gradient-based optimization algorithms for accelerating or improving training deep neural networks (DNNs). In deep learning practice, the momentum is usually weighted by a well-calibrated constant. However, tuning the hyperparameter for momentum can be a significant computational burden. In this article, we propose a novel adaptive momentum for improving DNNs training; this adaptive momentum, with no momentum-related hyperparame- ter required, is motivated by the nonlinear conjugate gradient (NCG) method. Stochastic gradient descent (SGD) with this new adaptive momentum eliminates the need for the momentum hyperparameter calibration, allows using a significantly larger learning rate, accelerates DNN training, and improves the final accuracy and robustness of the trained DNNs. For example, SGD with this adaptive momentum reduces classification errors for training ResNet110 for CIFAR10 and CIFAR100 from 5.25% to 4.64% and 23.75% to 20.03%, respectively. Furthermore, SGD, with the new adaptive momentum, also benefits adversarial training and, hence, improves the adversarial robustness of the trained DNNs.
Turbine performance directly affects engine specific impulse, thrust-to-weight ratio, and cost in a rocket propulsion system. A global optimization framework combining the radial basis neural network (RBNN) and the polynomial-based response surface method (RSM) is constructed for shape optimization of a supersonic turbine. Based on the optimized preliminary design, shape optimization is performed for the first vane and blade of a 2-stage supersonic turbine, involving O(10) design variables. The design of experiment approach is adopted to reduce the data size needed by the optimization task. It is demonstrated that a major merit of the global optimization approach is that it enables one to adaptively revise the design space to perform multiple optimization cycles. This benefit is realized when an optimal design approaches the boundary of a pre-defined design space. Furthermore, by inspecting the influence of each design variable, one can also gain insight into the existence of multiple design choices and select the optimum design based on other factors such as stress and materials considerations.
Here, we demonstrate a data-driven approach to interpreting surface reactions by combining time-resolved gas pulsing infrared spectroscopy with chemical reaction neural networks (CRNNs). Using CO adsorption and desorption on Pd(111) at 460–490 K as a model system, we show how transient kinetic data can reveal detailed reaction mechanisms. Starting with a simple one-species model, we systematically evaluate increasingly complex mechanisms involving hollow and bridge site adsorption. Despite the similar goodness of fit to the same experimental absorbance data, our models predict distinct coverage dynamics for different adsorption sites. Through analysis of spectral peak stability and predicted dynamics, we identify a mechanism in which CO primarily adsorbs on bridge sites followed by rapid conversion to hollow sites as being the most physically consistent with experimental observations. This work provides a framework for extracting mechanistic insights from limited experimental data, demonstrating how machine learning can bridge the gap between transient kinetic measurements and a molecular-level understanding of surface reactions.
Deep learning techniques have been useful for modern microscopy imaging techniques to further study and analyze biological structures and organs. Convolutional neural networks (CNN) have improved 2D object detection, localization, and segmentation. For imagery containing biological structures with depth, it is especially desirable to perform these tasks in 3D. Traditionally, performing these tasks simultaneously in 3D has proven to be computationally expensive. Currently available methodologies thus largely work to segment 3D objects from 2D images (without context from captured 3D volumes). In this work, we present a novel approach to perform fast and accurate localization, detection, and segmentation of volumes containing cells. Specifically, in our method, we modify and tune two state-of-the-art CNNs, namely 2D YOLOv2 and 3D U-Net, and combine them with a new fusion and image processing algorithms. Annotated volumes in this space are limited, and we have created synthetic data that mimics actual structures for training and testing our proposed approach. Promising results on this test data demonstrate the value of the technique and offers a methodology for 3D cell analysis in real microscopy imagery.
Damage initiation and localization in uniaxial tension tests of an Al6061-T6 alloy are evaluated and quantified using in situ X-ray computed tomography and a neural network segmentation strategy. This segmentation strategy offers significant advantages over traditional thresholding in quantifying microstructural and damage evolution metrics. Void nucleation was observed to occur throughout the deformation process, whereas void volume did not grow significantly until later stages of deformation. Localized analysis determined that void nucleation begins in and around the necked region during the early stages of deformation, whereas limited damage nucleation occurs outside of the neck until late stages of deformation. Lastly, inhomogeneities in void and particle spacing in the initial undeformed state of the material are observed to correlate well with the location of damage localization and necking during deformation. These results provide new insights into the ductile failure of Al alloys and how preexisting defects influence damage and rupture.
Researchers from Sandia National Laboratories (Sandia) and the University of Texas at Austin (UT) conducted this study to explore the effectiveness of commercial artificial neural network (ANN) software to improve insider threat detection and mitigation (ITDM). This study hypothesized that ANNs could be "trainee to learn patterns of organizational behaviors, detect off-normal (or anomalous) deviations from these patterns, and alert when certain types, frequencies, or quantities of deviations emerge. The ReconaSense ANN system was installed at UT's Nuclear Engineering Teaching Laboratory (NETL) and collected 13,653 access control data points and 694 intrusion sensor data points over a three-month period. Preliminary analysis of this baseline data demonstrated regularized patterns of life in the facility, and that off-normal behaviors are detectable under certain situations -- even for a facility with anticipated highly non-routine, operational behaviors. Completion of this pilot study demonstrated how the ReconaSense ANN could be used to identify expected operational patterns and detect unexpected anomalous behaviors in support of a data-analytic approach to ITDM. While additional studies are needed to fully understand and characterize this system, the results of this initial study are overall very promising for demonstrating a new framework for ITDM utilizing ANNs and data analysis techniques.
Full-physics cosmological simulations are powerful tools for studying the formation and evolution of structure in the universe but require extreme computational resources. Here, we train a convolutional neural network to use a cheaper N-body-only simulation to reconstruct the baryon hydrodynamic variables (density, temperature, and velocity) on scales relevant to the Lyα forest, using data from Nyx simulations. We show that our method enables rapid estimation of these fields at a resolution of ~20 kpc, and captures the statistics of the Lyα forest with much greater accuracy than existing approximations. Because our model is fully convolutional, we can train on smaller simulation boxes and deploy on much larger ones, enabling substantial computational savings. Furthermore, as our method produces an approximation for the hydrodynamic fields instead of Lyα flux directly, it is not limited to a particular choice of ionizing background or mean transmitted flux.
We present GFlaT, a new algorithm that uses a graph-neural-network to determine the flavor of neutral 𝐵 mesons produced in ϒ(4𝑆) decays. It improves previous algorithms by using the information from all charged final-state particles and the relations between them. We evaluate its performance using 𝐵 decays to flavor-specific hadronic final states reconstructed in a 362 fb −1 sample of electron-positron collisions collected at the ϒ(4𝑆) resonance with the Belle II detector at the SuperKEKB collider. We achieve an effective tagging efficiency of (37.40 ± 0.43 ± 0.36%), where the first uncertainty is statistical and the second systematic, which is 18% better than the previous Belle II algorithm. Demonstrating the algorithm, we use 𝐵 0 →𝐽/𝜓𝐾$^0_ S$ decays to measure the mixing-induced and direct 𝐶𝑃 violation parameters, 𝑆 = (0.724 ± 0.035 ± 0.009) and 𝐶 = (−0.035 ± 0.026 ± 0.029).
Deep reinforcement learning (DRL) has empowered a variety of artificial intelligence fields, including pattern recognition, robotics, recommendation-systems, and gaming. Similarly, graph neural networks (GNN) have also demonstrated their superior performance in supervised learning for graph-structured data. In recent times, the fusion of GNN with DRL for graph-structured environments has attracted a lot of attention. Here, this paper provides a comprehensive review of these hybrid works. These works can be classified into two categories: (1) algorithmic enhancement, where DRL and GNN complement each other for better utility; (2) application-specific enhancement, where DRL and GNN support each other. This fusion effectively addresses various complex problems in engineering and life sciences. Based on the review, we further analyze the applicability and benefits of fusing these two domains, especially in terms of increasing generalizability and reducing computational complexity. Finally, the key challenges in integrating DRL and GNN, and potential future research directions are highlighted, which will be of interest to the broader machine learning community.
Paper information. Shuang Yu, Indrasis Chakraborty, Gemma J. Anderson, Donald D. Lucas, Yannic Lops, and Daniel Galea. UFNet: Joint U-Net and fully connected neural network to bias correct precipitation predictions from climate models. Artificial Intelligence for the Earth Systems, 2024. Overview. This work develops the UFNet methodology to correct E3SM historical precipitation projection bias. The UFNet deep learning framework consists of a two-part architecture: a U-Net convolutional network to capture the spatiotemporal distribution of precipitation and a fully connected network to capture the distribution of higher-order statistics. The joint network, termed UFNet, can simultaneously improve the spatial structure of the modeled precipitation and capture the distribution of extreme precipitation values. Below we provide guidance for applying UFNet to correct the Energy Exascale Earth System Model (E3SM; Golaz et al. 2019) daily precipitation projection over the contiguous United States (CONUS). Getting started 1. Obtain the historical climate simulation and observation data. The E3SM historical simulation data are available through https://aims2.llnl.gov/search/cmip6/. The CPC unified gauge-based analysis of daily precipitation can be found through https://psl.noaa.gov/data/gridded/data.cpc.globalprecip.html. The ECMWF atmospheric reanalysis of the 20th century (ERA-20C) data are available through https://www.ecmwf.int/en/forecasts/datasets/reanalysis-datasets/era-20c. The spatial resolution of E3SM and observed datasets are both regridded to a common 1° resolution grid using conservative interpolation. The regridded E3SM, CPC and ERA-20C with 1° resolution can be found throught ./data/. 2. Train the fully connected network (DNN) Python train_dnn.py 3. Train the UFNet Python train_ufnet.py 4. Evaluation and compared with the baseline Python evaluation.py
Real-time analysis of patient data during medical procedures can provide vital diagnostic feedback that significantly improves chances of success. With sensors becoming increasingly fast, frameworks such as Deep Neural Networks are required to perform calculations within the strict timing constraints for real-time operation. However, traditional computing platforms responsible for running these algorithms incur a large overhead due to communication protocols, memory accesses, and static (often generic) architectures. In this work, we implement a low-latency Multi-Layer Perceptron (MLP) processor using Field Programmable Gate Arrays (FPGAs). Unlike CPUs and Graphics Processing Units (GPUs), our FPGA-based design can directly interface sensors, storage devices, display devices and even actuators, thus reducing the delays of data movement between ports and compute pipelines. Moreover, the compute pipelines themselves are tailored specifically to the application, improving resource utilization and reducing idle cycles. We demonstrate the effectiveness of our approach using mass-spectrometry data sets for real-time cancer detection. We demonstrate that correct parameter sizing, based on the application, can reduce latency by 20% on average. Furthermore, we show that in an application with tightly coupled data-path and latency constraints, having a large amount of computing resources can actually reduce performance. Using mass-spectrometry benchmarks, we show that our proposed FPGA design outperforms both CPU and GPU implementations, with an average speedup of 144x and 21x, respectively. In our work, we demonstrate the importance of application-specific optimizations in order to minimize latency and maximize resource utilization for MLP inference. By directly interfacing and processing sensor data with ultra-low latency, FPGAs can perform real-time analysis during procedures and provide diagnostic feedback that can be critical to achieving higher percentages of successful patient outcomes.
Quantum information processing offers potentially faster performance compared to the classical counterpart. The goal of this work is to investigate performance of Quantum Hopfield Neural Network for applications to anomaly detection. Preliminary study consist of assessment of computational capabilities of Qiskit quantum simulator. Eventual objective is to implement Quantum Hopfield Network algorithm for detection of weak nuisance and anomaly signal in the presence of strong and highly varying background in gamma radiation data measured during environmental screening.