SEARCH · Engineering Papers
Results for “physical learning”
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.
Extension of the PINN diffusion model to k-eigenvalue problems
This paper extends our recent work on the Physics-Informed Neural Networks (PINN) approach for the fixed source diffusion models and applies it to the diffusion theory based k-eigenvalue problems. To make the PINN equitable for the eigenvalue problems, we introduce a novel integral regularization term to the loss function in the framework, and allow the direct inference of the principal eigenvalue and the associated eigenfunction. The regularization term enforces a pre-defined value on the integration of the model predictions, and this value can be directly related to a physical property of the system. We also introduce an additional learnable parameter to approximate the principal eigenvalue. As a proof of principle, we solve the one-group two-dimensional k-eigenvalue neutron diffusion equation in this work. We then provide two numerical examples to demonstrate the applicability of the PINN approach. In each example, we solve the k-eigenvalue diffusion equation in a multi-region configuration constrained with a set of Robin boundary conditions for generality. We use a FEM solution based on the power-iteration method to verify the results of the PINN solution. The results showed relative percentage error in the predicted eigenvalue of about 0.77% and about 1.2% for example 1 and example 2, respectively. The mean absolute error in the predicted flux for example 1 is ∼ 0.002 and for example 2 is ∼ 0.0024. These results indicate some preliminary successes of the PINN application to k-eigenvalue problems. (authors)
Proxy Applications for Converged Workloads: DMC LDRD Initiative
Modern scientific applications are complicated and require coordination of several components. Proxy application driven software-hardware co-design plays a vital role in driving innovation among the developments of applications, software infrastructure and hardware architecture. Proxy applications are self-contained and simplified codes that are intended to model the performance-critical computations within applications. Applications executing on modern High Performance Computing (HPC) systems are susceptible to network congestion, insufficient memory bandwidth within and across compute nodes, and inadvertent loss of performance due to bugs and unoptimized programming models. Modern numerical simulations and machine learning models play a critical role in studying physical phenomenon under myriad uncertainties. Such applications often exhibit irregular computation and memory accesses at specific regions of the application code, which can contribute to various performance bottlenecks at scale. To mitigate such issues and prepare the next generation hardware for a variety of computation and data movement contingencies, a well-known practice is to consider "proxy" applications as representative motifs for various classes of scientific applications. While there is disagreement in the HPC community on the mechanisms of construction of the proxy applications, there is a strong consensus on their positive impact in co-design. Proxy Applications for Converged Workloads (PACER) is about facilitating software-hardware co-design through proxy applications with the goal of improving the performance of converged science workflows on heterogeneous systems.
Predicting Catalyst Surface Stability Under Reaction Conditions Using Deep Reinforcement Learning and Machine Learning Potentials
Catalysts are critical for most large-scale energy intensive chemical transformation processes, such as energy storage, liquid fuel production, and the formation of chemical building blocks. The catalyst composition, structure, and morphology impact the performance under reaction conditions and influences properties like activity and selectivity. Many industrial catalysts offer imperfect activity/selectivity, or contain expensive metals. Further, catalysts can deactivate over time as the morphology changes or harsh reactive environments alter the surface structure. Methods to automatically model and predict the kinetics of how catalyst surfaces will restructure would enable engineers to design around these challenges, improve performance, increase longevity. This project investigated a specific type of machine learning model, deep reinforcement learning, machine learning models to act as surrogates for the physical system, and compared the results of these approaches with a specialized high-throughput experimental synthesis and measurement platform.
Applications and Techniques for Fast Machine Learning in Science
In this community review report, we discuss applications and techniques for fast machine learning (ML) in science—the concept of integrating powerful ML methods into the real-time experimental data processing loop to accelerate scientific discovery. The material for the report builds on two workshops held by the Fast ML for Science community and covers three main areas: applications for fast ML across a number of scientific domains; techniques for training and implementing performant and resource-efficient ML algorithms; and computing architectures, platforms, and technologies for deploying these algorithms. We also present overlapping challenges across the multiple scientific domains where common solutions can be found. This community report is intended to give plenty of examples and inspiration for scientific discovery through integrated and accelerated ML solutions. This is followed by a high-level overview and organization of technical advances, including an abundance of pointers to source material, which can enable these breakthroughs.
Efficient solutions of fermionic systems using artificial neural networks
In this study, we explore the similarities and differences between variational Monte Carlo techniques that employ conventional and artificial neural network representations of the ground-state wave function for fermionic systems. Our primary focus is on shallow neural network architectures, specifically the restricted Boltzmann machine, and we examine unsupervised learning algorithms that are appropriate for modeling complex many-body correlations. We assess the advantages and drawbacks of conventional and neural network wave functions by applying them to a range of circular quantum dot systems. Our findings, which include results for systems containing up to 90 electrons, emphasize the efficient implementation of these methods on both homogeneous and heterogeneous high-performance computing facilities.
Growth and properties of Hg-based quantum well structures and superlattices
An overview of the properties of HgTe-CdTe quantum well structures and superlattices (SL) is presented. These new quantum structures are candidates for use as new long wavelength infrared (LWIR) and very long wavelength infrared (VLWIR) detectors, as well as for other optoelectronic applications. Much has been learned within the past two years about the physics of such structures. The valence band offset has been determined to be approx. 350 meV, independent of temperature. The occurrence of electron and hole mobilities in excess of 10(exp 5)cm(exp 2)/V center dot s is now understood on the basis of SL band structure calculations. The in-plane and out-of-plane electron and hole effective masses have been measured and interpreted theoretically for HgTe-CdTe superlattices. Controlled substitutional doping of superlattices has recently been achieved at North Carolina State University (NCSU), and modulation-doped SLs have now been successfully grown and studied. Most recently, a dramatic lowering of the growth temperature of Hg-based quantum well structure and SLs (to approx. 100 C) has been achieved by means of photoassisted molecular beam epitaxy (MBE) at NCSU. A number of new devices have been fabricated from these doped multilayers.
Modeling reality
Although powerful computers have allowed complex physical and manmade hardware systems to be modeled successfully, we have encountered persistent problems with the reliability of computer models for systems involving human learning, human action, and human organizations. This is not a misfortune; unlike physical and manmade systems, human systems do not operate under a fixed set of laws. The rules governing the actions allowable in the system can be changed without warning at any moment, and can evolve over time. That the governing laws are inherently unpredictable raises serious questions about the reliability of models when applied to human situations. In these domains, computers are better used, not for prediction and planning, but for aiding humans. Examples are systems that help humans speculate about possible futures, offer advice about possible actions in a domain, systems that gather information from the networks, and systems that track and support work flows in organizations.
Extreme Ultraviolet Explorer. Long look at the next window
The Extreme Ultraviolet Explorer (EUVE) will map the entire sky to determine the existence, direction, brightness, and temperature of thousands of objects that are sources of so-called extreme ultraviolet (EUV) radiation. The EUV spectral region is located between the x-ray and ultraviolet regions of the electromagnetic spectrum. From the sky survey by EUVE, astronomers will determine the nature of sources of EUV light in our galaxy, and infer the distribution of interstellar gas for hundreds of light years around the solar system. It is from this gas and the accompanying dust in space that new stars and solar systems are born and to which evolving and dying stars return much of their material in an endless cosmic cycle of birth, death, and rebirth. Besides surveying the sky, astronomers will make detailed studies of selected objects with EUVE to determine their physical properties and chemical compositions. Also, they will learn about the conditions that prevail and the processes at work in stars, planets, and other sources of EUV radiation, maybe even quasars. The EUVE mission and instruments are described. The objects that EUVE will likely find are described.
Mineralogy of the Solar System
The coming decade will witnesses the first sample return missions from solar system bodies in 30 years. These samples will all be very small, some missions return only a few milligrams of total mass. Fortunately, the capability of modem methods to characterize ultra-small samples is well established from analysis of interplanetary dust particles (IDPs), interstellar grains recovered from meteorites, and other materials requiring ultra-sensitive analytical capabilities. Powerful analytical techniques are available that require, under favorable circumstances, single particles of only a few nanograms for entire suites of fairly comprehensive characterizations. A returned sample of greater than 1,000 particles with total mass of just one microgram permits comprehensive quantitative geochemical measurements that are impractical to can-y out in situ by flight instruments. With the Galileo flybys of Gaspra and Ida, it is now recognized that even very small airless bodies have indeed developed a particulate regolith. Acquiring a sample of the bulk regolith, a simple sampling strategy, provides two critical pieces of information about the body. Regolith samples are excellent bulk samples since they normally contain all the key components of the local environment, albeit in particulate form. Furthermore, since this fine fraction dominates remote measurements, regolith samples also provide information about surface alteration processes and are a key link to remote sensing of other bodies. Studies indicate that a statistically significant number of nanogram-sized particles should be able to characterize the regolith of a primitive asteroid, although the presence of larger components within even primitive meteorites (e.g.. Murchison), e.g. chondrules, CAI, large crystal fragments, etc., points out the limitations of using data obtained from nanogram-sized samples to characterize entire primitive asteroids. However, most important asteroidal geological processes have left their mark on the matrix, since this is the finest-grained portion and therefore most sensitive to chemical and physical changes. Thus, the following information can be learned from this fine grain size fraction alone: (1) mineral paragenesis; (2) regolith processes, (3) bulk composition; (4) conditions of thermal and aqueous alteration (if any); (5) relationships to planets, comets, meteorites (via isotopic analyses, including oxygen; (6) abundance of water and hydrated material; (7) abundance of organics; (8) history of volatile mobility, (9) presence and origin of presolar and/or interstellar material.
Hybrid data-driven and model-informed online tool wear detection in milling machines
Precision machining tool wear is responsible for low product throughput and quality. Monitoring the tool wear online is vital to prevent degradation in machining quality. However, direct real-time tool wear measurement is not practical. This paper presents residual-based anomaly detection models, combining a hybrid model comprised of a physics-based model and a data-driven model (a decision tree or a neural network) to predict signals of interest (e.g., power or forces) under nominal conditions, followed by Page’s cumulative sum test for detecting tool wear on-line using the computer numerical control machine measurements. The most informative features are ranked using dynamic programming and its approximation variants from real-time measurements and machine settings, such as the width of cut, depth of cut, feed rate and spindle speed, that serve as inputs to the predictive models. The baseline nominal model is incrementally updated with experimental data via a gradient boosted adaptation model to generate the residuals that account for discrepancies between the actual machine data under normal conditions and the baseline nominal model predictions. The hybrid model is validated against 20 Mazak milling machine experimental tests and one Haas run-to-failure experiment. The proposed anomaly detector is applied to synthetic data from simulations of the physics-based model at different operating conditions, measurement noise levels, and tool wear levels, and the methods were able to achieve an overall 92% accuracy in data with 1% noise. The anomaly detection methods based on hybrid model reduced the false alarms of either the data-driven or physical-based models alone, and are found to be capable of good online detection of tool wear.
AmesDT: Digital Twin and Autonomy Validation Environment
A simulation of NASA Ames Research Center was developed to provide a common testbed for multiple areas of research within the Intelligent Systems Division, primarily related to verification and validation of autonomous technologies, machine learning, and digital twin systems. AmesSim corresponds a physical rover that is capable of navigation in the real-world environment; in this way, the same experiments can be run in both settings, with the same software and hardware stacks in the loop. The simulation is built in Unreal Engine 4 and uses the AirSim plugin for API convenience. Several custom modifications allow deterministic, faster-than-realtime execution, which enables consistent testing of on-line algorithms and large-scale data collection. This paper describes the architecture and capabilities of the simulation and discusses development challenge.
Peak Prediction Using Multi Layer Perceptron (MLP) for Edge Computing ASICs Targeting Scientific Applications
High data rate detectors play an integral part in scientific research and their development is actively pursued at High Energy Physics (HEP) facilities around the world. Edge Machine Learning (ML) offers the ability to reduce data rates by integrating ML algorithms into Application Specific Integrated Circuits (ASICs) on the front end electronics. In this work, we explore a set of neural network architectures for predicting the peak amplitudes in the detector's sensor response. We have designed and synthesized several MLP based neural networks comparing their inference accuracy, power consumption, and area targeting for minimal latency. The neural networks are synthesized in a commercial 65nm process. The effect of quantizing the network's weights and biases on hardware performance and area is reported. We also conduct design space exploration to compare between design alternatives
Embedding Neural Thermal Scattering (NeTS) Modules in SERPENT for Higher Fidelity Advanced Reactor Analysis
When a neutron born in fission thermalizes to the order of $k$ $B$ $T$, it’s de-Broglie wavelength and energy approach the order of inter-atomic spacing and elementary lattice oscillations, respectively. $S$($a,β,t$) or the scattering law, uuantify these temperature-dependent crystallographic contributions to total cross section (or reaction rate). In a Monte Carlo analysis, cumulative distribution functions (CDFs) of $S$($a,β,t$) are loaded to memory from “A Compact ENDF” (ACE) files for stochastically selecting thermal scattered neutron trajectories. In this work, novel neural thermal scattering (NeTS) modules for $S$($a,β,t$) CDFs are designed, trained, serialized and embedded within SERPENT using Python’s limited C-API for on-the-fly deployment of crystalline graphite $S$($a,β,t$) sampling. Torchscript tracing and Numba just-in-time (JIT) compilation streamline neural inference on NVIDIA GPUs with CUDA libraries. Demonstrations of bare sphere thermalization of fast and thermal sources show excellent agreement between embedded NeTS in SERPENT and MCNP. With an explicit model of the reactor, NeTS can predict on-the-fly changes in TREAT neutron spectra as a function of local temperature, which can serve to improve transient and accident predictions in a multiphysics analysis framework. This framework can be further extended to account on-the-fly for changes in local graphitic microstructure to scattering cross sections, and outlines a novel coupling of modern machine learning with state-of-the-art reactor physics methods.
Engagement: Hyperparameter Optimization of Generative Adversarial Network Models for High-Energy Physics Simulations
We present our SciDAC FASTMath-HEP partnership results for tuning generative adversarial models (GANs) for high energy physics applications. The GANs are used in hybrid simulations to accelerate otherwise time-consuming computations. We optimize for both, prediction accuracy and variability with the goal to find GAN architectures that are reliable and robust.
Precise Modeling of a Complex Solenoidal Magnetic Field Using a Combination of Analytic Functions and a PINN
We demonstrate an iterative approach to modeling a sparsely measured magnetic field in a large-bore solenoid. This approach uses a hybrid of traditional and machine learning techniques. The traditional technique is a linear least-squares fit using a series solution to Laplace's equation, while the machine learning technique involves the training of a physics-informed neural network (PINN) on the least-squares fit residuals. We use a newly defined activation function "DELTAsnake," a modification to the snake activation function proposed by Ziyin et al. that allows for stronger curvature and non-monotonicity. The combined model approximately obeys Maxwell's equations to a level sufficient for producing high quality physics simulations and analysis. Our approach is applied to a highly realistic calculation of the expected magnetic field in the Mu2e experiment's Detector Solenoid which includes a simple model for the expected statistical measurement uncertainties. Using ten toy measurement simulations, we demonstrate the capabilities of our model in comparison to the least-squares method alone; the least-squares method alone results in a reduced chi-squared statistic of ${2.15 \pm 0.01}$, while our approach improves the reduced chi-square to ${1.034 \pm 0.005}$. Furthermore, for an average toy simulation, we show that the range of the RMS of the three field component residuals reduces from ${0.07-0.37}$ Gauss to ${0.05-0.07}$ Gauss. We find that this novel method is robust against a realistic systematic uncertainty deriving from Hall probe calibration bias and can be used to significantly reduce the number of measurements required to achieve an accurate model.
Data compression and inference in cosmology with self-supervised machine learning
ABSTRACT The influx of massive amounts of data from current and upcoming cosmological surveys necessitates compression schemes that can efficiently summarize the data with minimal loss of information. We introduce a method that leverages the paradigm of self-supervised machine learning in a novel manner to construct representative summaries of massive data sets using simulation-based augmentations. Deploying the method on hydrodynamical cosmological simulations, we show that it can deliver highly informative summaries, which can be used for a variety of downstream tasks, including precise and accurate parameter inference. We demonstrate how this paradigm can be used to construct summary representations that are insensitive to prescribed systematic effects, such as the influence of baryonic physics. Our results indicate that self-supervised machine learning techniques offer a promising new approach for compression of cosmological data as well as its analysis.
Bio-Inspired Neural Model for Learning Dynamic Models
A neural-network mathematical model that, relative to prior such models, places greater emphasis on some of the temporal aspects of real neural physical processes, has been proposed as a basis for massively parallel, distributed algorithms that learn dynamic models of possibly complex external processes by means of learning rules that are local in space and time. The algorithms could be made to perform such functions as recognition and prediction of words in speech and of objects depicted in video images. The approach embodied in this model is said to be "hardware-friendly" in the following sense: The algorithms would be amenable to execution by special-purpose computers implemented as very-large-scale integrated (VLSI) circuits that would operate at relatively high speeds and low power demands.