Unsupervised Impedance and Topology Estimation of Distribution Networks—Limitations and Tools
Not provided.
Engineering↗
SEARCH · Search NASA
Results for “Network modeling”
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Two-component modeling of the solar IR CO lines
One-dimensional hydrostatic models of quiet and active solar regions can be constructed that generally account for the observed intensities of lines and continue throughout the spectrum, except for the infrared CO lines. There is an apparent conflict between: (1) observations of the strongest infrared CO lines formed in LTE at low-chromospheric heights but at temperatures much cooler than the average chromospheric values; and (2) observations of Ca II, UV (ultraviolet), and microwave intensities that originate from the same chromospheric heights but at the much higher temperatures characteristic of the average chromosphere. A model M(sub CO) has been constructed which gives a good fit to the full range of mean CO line profiles (averaged over the central area of the solar disk and over time) but this model conflicts with other observations of average quiet regions. A model L(sub CO) which is approximately 100 K cooler than M(sub CO) combined with a very bright network model F in the proportions 0.6 L(sub CO) + 0.4 F is found to be generally consistent with the CO, Ca II, UV, and microwave observations. Ayres, Testerman, and Brault found that models COOLC and FLUXT in the proportions 0.925 and 0.075 account for the CO and Ca II lines, but these combined models give an average UV intensity at 140 nm about 20 times larger than observed. The 0.6 L(sub CO) + 0.4 F result may give a better description of the cool and hot components that produce the space- and time-averaged spectra. Recent observations carried out by Uitenbroek, Noyes, and Rabine with high spatial and temporal resolution indicate that the faintest intensities in the strong CO lines measured at given locations usually become much brighter within 1 to 3 minutes. The cool regions thus seem to be mostly the low-temperature portions of oscillatory waves rather than cool structures that are stationary.
Avrett, Eugene H.↗
Modeling the 4D discharge of lithium-ion batteries with a multiscale time-dependent deep learning framework
The lithium-ion battery (LIB) field is moving towards the direction of investigating spatially resolved physical phenomena in the 3D porous microstructure of electrodes. These pore-scale simulations give new insights into the local dynamics of lithiation/de-lithiation and charge transport, Nevertheless, the computational time of these simulations limits the integration of these models in optimization workflows of cycling conditions or electrode manufacturing processes. Machine learning models present a way of assessing in real-time the performance of materials. While several successful techniques for replicating simulations with machine learning have been proposed, this case study presents a more demanding problem, due to the necessity of understanding the behavior of heterogeneous 3D local data, as it evolves in time: this poses both a scientific and a technical challenge. To this end, we propose an autoregressive multiscale convolutional neural network model to predict relevant quantities at the pore-scale in the solid phase: the lithium concentration (in the active material) and potential (in the active material and carbon binder). Here, these are ultimately used to reconstruct the battery discharge curve. 3D images of the electrode microstructures are the input to the network, trained with a dataset of finite element method simulations to predict the discharge behavior of the cathode side in lithium ion batteries. We propose this machine learning model as a proof-of-concept of the applicability of multiscale networks for time-dependent physics problems. The trained model exhibits very high accuracy (with errors lower than 2 %) in forecasting the discharge behavior of new unseen cathodes.
25 ENERGY STORAGE↗
Architecture for Cognitive Networking within NASAs Future Space Communications Infrastructure
Future space mission concepts and designs pose many networking challenges for command, telemetry, and science data applications with diverse end-to-end data delivery needs. For future end-to-end architecture designs, a key challenge is meeting expected application quality of service requirements for multiple simultaneous mission data flows with options to use diverse onboard local data buses, commercial ground networks, and multiple satellite relay constellations in LEO, MEO, GEO, or even deep space relay links. Effectively utilizing a complex network topology requires orchestration and direction that spans the many discrete, individually addressable computer systems, which cause them to act in concert to achieve the overall network goals. The system must be intelligent enough to not only function under nominal conditions, but also adapt to unexpected situations, and reorganize or adapt to perform roles not originally intended for the system or explicitly programmed. This paper describes architecture features of cognitive networking within the future NASA space communications infrastructure, and interacting with the legacy systems and infrastructure in the meantime. The paper begins by discussing the need for increased automation, including inter-system collaboration. This discussion motivates the features of an architecture including cognitive networking for future missions and relays, interoperating with both existing endpoint-based networking models and emerging information-centric models. From this basis, we discuss progress on a proof-of-concept implementation of this architecture as a cognitive networking on-orbit application on the SCaN Testbed attached to the International Space Station.
space networks↗
Lithium-Ion Battery Diagnostics Using Electrochemical Impedance via Machine-Learning
Diagnosing battery states such as health, state-of-charge, or temperature is crucial for ensuring the safety and reliability of electrochemical energy storage systems. While some states, such as temperature, may be measured using cheap sensors, accurate diagnosis of battery health metrics usually requires time-consuming performance measurements, making them infeasible for use in real-world operation. These health metrics can be measured during lab-testing and then estimated on-line using predictive life models or via state observer algorithms such as Kalman filters, but these predictive methods should be supplemented by actual measurement of battery health whenever possible to ensure reliability. Rapid measurement of battery health may be done by various types of fast diagnostic techniques such as electrochemical impedance spectroscopy (EIS), which can be performed in only a few minutes and require only a fraction of the energy and power needed for a full charge and discharge measurement. But there is a substantial challenge for estimating battery health using EIS data, as EIS is sensitive to cell temperature, state-of-charge, current, and resting time in addition to health. Thus, utilizing EIS data to predict battery capacity requires correcting for all these additional variables, a task that is extremely difficult to handle analytically. This talk utilizes machine-learning methods to estimate the effectiveness of battery capacity prediction from EIS data, leveraging a data set of hundreds of EIS measurements recorded at varying temperature and state-of-charge throughout a 500-day aging study of 32 commercial, large-format NMC-Graphite lithium-ion batteries. Using EIS as input to machine-learning models is complicated by the nonlinear response of impedance to battery health, temperature, and state-of-charge, as well as the collinearity between the impedance response at neighboring frequencies, which can easily lead to overfit models. To train robust models, features from EIS data need to be extracted from the data or some subset of critical frequencies selected. Many approaches for extracting and selecting features from EIS data from electrochemical analysis and machine-learning fields were identified for analysis: using the entire raw spectra; selection of one, two, or many frequencies from the entire spectra; selecting interesting points from the EIS measurement using domain knowledge; fitting EIS with an equivalent-circuit model; calculating statistics on the raw impedance values; and reducing the dimensionality of the data using unsupervised linear (principal component analysis) and non-linear (uniform manifold approximation and projection) methods. These approaches were rigorously compared using a machine-learning pipeline approach, training linear, Gaussian process, and random forest regression models and quantifying performance using cross-validation as well as a held-out test set. An artificial neural network model trained on the raw spectra was also tested. Promising pipelines were fine-tuned via Bayesian hyperparameter optimization using cross-validation loss and training with class-specific weights to counter data set imbalance. The most reliable method for utilizing impedance in this work was the selection of two optimal frequencies through an exhaustive search, resulting in about 2% mean absolute error on test data for both Gaussian process and random forest model architectures. Interrogation of a variety of models reveals critical frequencies of 100 Hz and 103 Hz for this data set, though the optimal set of frequencies is not necessarily intuitive, i.e., the best performing models are not simply those that use impedance at frequencies that have the highest correlation to the relative discharge capacity. The best performing model is an ensemble model, which is able to predict battery capacity with 1.9% mean absolute error for unseen cells using impedance recorded at a variety of temperatures and states-of-charge.
battery↗
Applying Machine Learning to Predict Alaskan Ionospheric Irregularities
In this work several machine-learning (ML) techniques for predicting ionospheric irregularities in the northern auroral zone were tested. The techniques include Ridge Regression, Long Short-Term Memory Neural Network (LSTM), Classification Neural Network (CNN), Autoencoder Classification Neural Network (ACNN), and LSTM Autoencoder Classification Neural Network (LACNN). These techniques were tested with the rate of total electron content (TEC) index (ROTI) data collected during 2008 and 2009 from a geodetic station in Fairbanks, Alaska (64.98°N, 147.50°W), which is in the auroral zone. Using ROTI data with the ML techniques, experiments were conducted to reach two goals: (1) examine what space weather measurements present good correlation with ROTI so that they may be helpful in ML-based prediction of ionospheric irregularities in the polar region; (2) predict ROTI hours and days ahead by training the neural network models with historical ROTI data alone. The Ridge Regression experiments indicate that a combination of measurements of local geomagnetic horizontal components, geomagnetic SYM-H index, 3-hour Kp and ap indices, and F10.7 solar flux index appears to be more correlated to the single-site ROTI measurements than other parameters. The neural network (NN) experiments show that although the LACNN model allows for predictions of non-irregularity and irregularity conditions defined by ROTI levels up to 3 hours in advance, with an overall accuracy ≥ 92%, a number of irregularity events can still be missed. Hence, further development is needed to reduce the number of missed events. In this paper, the models, data processing, model performance, prediction results, and potential applications are presented.
Pi, Xiaoqing↗
Solving sparse finite element problems on neuromorphic hardware
The finite element method (FEM) is one of the most important and ubiquitous numerical methods for solving partial differential equations (PDEs) on computers for scientific and engineering discovery. Applying the FEM to larger and more detailed scientific models has driven advances in high-performance computing for decades. Here we demonstrate that scalable spiking neuromorphic hardware can directly implement the FEM by constructing a spiking neural network that solves the large, sparse, linear systems of equations at the core of the FEM. We show that for the Poisson equation, a fundamental PDE in science and engineering, our neural circuit achieves meaningful levels of numerical accuracy and close to ideal scaling on modern, inherently parallel and energy-efficient neuromorphic hardware, specifically Intel’s Loihi 2 neuromorphic platform. We illustrate extensions to irregular mesh geometries in both two and three dimensions as well as other PDEs such as linear elasticity. Our spiking neural network is constructed from a recurrent network model of the brain’s motor cortex and, in contrast to black-box deep artificial neural network-based methods for PDEs, directly translates the well-understood and trusted mathematics of the FEM to a natively spiking neuromorphic algorithm.
Applied mathematics↗
Predictability of Seawater DMS During the North Atlantic Aerosol and Marine Ecosystem Study (NAAMES)
This work presents an overview of a unique set of surface ocean dimethylsulfide (DMS) measurements from four shipboard field campaigns conducted during the North Atlantic Aerosol and Marine Ecosystem Study (NAAMES) project. Variations in surface seawater DMS are discussed in relation to biological and physical observations. Results are considered at a range of timescales (seasons to days) and spatial scales (regional to sub-mesoscale). Elevated DMS concentrations are generally associated with greater biological productivity, although chlorophyll a (Chl) only explains a small fraction of the DMS variability (15%). Physical factors that determine the location of oceanic temperature fronts and depth of vertical mixing have an important influence on seawater DMS concentrations during all seasons. The interplay of biomass and physics influences DMS concentrations at regional/seasonal scales and at smaller spatial and shorter temporal scales. Seawater DMS measurements are compared with the global seawater DMS climatology and predictions made using a recently published algorithm and by a neural network model. The climatology is successful at capturing the seasonal progression in average seawater DMS, but does not reproduce the shorter spatial/temporal scale variability. The input terms common to the algorithm and neural network approaches are biological (Chl) and physical (mixed layer depth, photosynthetically active radiation, seawater temperature). Both models predict the seasonal North Atlantic average seawater DMS trends better than the climatology. However, DMS concentrations tend to be under-predicted and the episodic occurrence of higher DMS concentrations is poorly predicted. The choice of climatological seawater DMS product makes a substantial impact on the estimated DMS flux into the North Atlantic atmosphere. These results suggest that additional input terms are needed to improve the predictive capability of current state-of-the-art approaches to estimating seawater DMS.
54 ENVIRONMENTAL SCIENCES↗
The Livermore Brain: Massive Deep Learning Networks Enabled by High Performance Computing
The proliferation of inexpensive sensor technologies like the ubiquitous digital image sensors has resulted in the collection and sharing of vast amounts of unsorted and unexploited raw data. Companies and governments who are able to collect and make sense of large datasets to help them make better decisions more rapidly will have a competitive advantage in the information era. Machine Learning technologies play a critical role for automating the data understanding process; however, to be maximally effective, useful intermediate representations of the data are required. These representations or “features” are transformations of the raw data into a form where patterns are more easily recognized. Recent breakthroughs in Deep Learning have made it possible to learn these features from large amounts of labeled data. The focus of this project is to develop and extend Deep Learning algorithms for learning features from vast amounts of unlabeled data and to develop the HPC neural network training platform to support the training of massive network models. This LDRD project succeeded in developing new unsupervised feature learning algorithms for images and video and created a scalable neural network training toolkit for HPC. Additionally, this LDRD helped create the world’s largest freely-available image and video dataset supporting open multimedia research and used this dataset for training our deep neural networks. This research helped LLNL capture several work-for-others (WFO) projects, attract new talent, and establish collaborations with leading academic and commercial partners. Finally, this project demonstrated the successful training of the largest unsupervised image neural network using HPC resources and helped establish LLNL leadership at the intersection of Machine Learning and HPC research.
97 MATHEMATICS AND COMPUTING↗
Gaussian Process Classification for Galaxy Blend Identification in LSST
Abstract A significant fraction of observed galaxies in the Rubin Observatory Legacy Survey of Space and Time (LSST) will overlap at least one other galaxy along the same line of sight, in a so-called “blend.” The current standard method of assessing blend likelihood in LSST images relies on counting up the number of intensity peaks in the smoothed image of a blend candidate, but the reliability of this procedure has not yet been comprehensively studied. Here we construct a realistic distribution of blended and unblended galaxies through high-fidelity simulations of LSST-like images, and from this we examine the blend classification accuracy of the standard peak-finding method. Furthermore, we develop a novel Gaussian process blend classifier model, and show that this classifier is competitive with both the peak finding method as well as with a convolutional neural network model. Finally, whereas the peak-finding method does not naturally assign probabilities to its classification estimates, the Gaussian process model does, and we show that the Gaussian process classification probabilities are generally reliable.
79 ASTRONOMY AND ASTROPHYSICS↗
Ion thruster thermal characteristics and performance.
Experimental and analytical results for a typical 20-cm-diam, hollow-cathode ion thrustor are reported. The foundation of the investigation was the application of thermal model correction techniques. Pertinent thermal properties and plasma heating characteristics of the thrustor were determined through correlation and integration of temperature measurement data with a single-state Wiener-Kalman filter. The thrustor self-heating levels on various parts were realistically estimated. Analytically predicted temperatures were forced to agree with the measured values for the purpose of constructing a corrected thermal model, which could then be used to evaluate more realistic thrustor circumstances and environments. The expected accuracy of the resultant analytical network model was demonstrated to be plus or minus 10 K. Thrustor thermal performance data for a typical five-thrustor array are presented as functions of environmental solar intensities. The thermal analyses are also extended to a 30-cm thrustor system.
Wen, L.↗
NEUROSPF: A Tool For the Symbolic Analysis of Neural Networks
This paper presents NEUROSPF, a tool for the symbolic analysis of neural networks. Given a trained neural network model, the tool extracts the architecture and model parameters and translates them into a Java representation that is amenable for analysis using the Symbolic PathFinder symbolic execution tool. Notably, NEUROSPF encodes specialized peer classes for parsing the model’s parameters, thereby enabling efficient analysis. With NEUROSPF the user has the flexibility to specify either the inputs or the network internal parameters as symbolic, promoting the application of program analysis and testing approaches from software engineering to the field of machine learning. For instance, NEUROSPF can be used for coverage-based testing and test generation, finding adversarial examples and also constraint-based repair of neural networks, thus improving the reliability of neural networks and of the applications that use them.
neural networks↗
Pilot Workload Rating Predictions Using Image Data and Recurrent Neural Networks
In this work, we augmented existing methods for estimating pilot workload ratings with deep neural networks trained using data from simulated flight tests in the Vertical Motion Simulator (VMS). We used an existing method, Spare Capacity Operations Estimator (SCOPE), along with a recurrent neural network and conducted comparison studies between the two methods individually, and when used together. We found that using both methods together can improve the result over using either approach alone. In our first test case, we achieved an improved linear correlation coefficient of 0.409 over that of SCOPE alone at 0.352 on the training dataset. Through cross validation, we also found that the results may be dependent on the split of training vs. validation data, and that further investigation should be conducted to understand what additional inputs to the neural network model should be made.
Image Data↗
Importance of Engineered and Learned Molecular Representations in Predicting Organic Reactivity, Selectivity, and Chemical Properties
Machine-readable chemical structure representations are foundational in all attempts to harness machine learning for the prediction of reactivities, selectivities, and chemical properties directly from molecular structure. The featurization of discrete chemical structures into a continuous vector space is a critical phase undertaken before model selection, and the development of new ways to quantitatively encode molecules is an active area of research. Here, we highlight the application and suitability of different representations, from expert-guided “engineered” descriptors to automatically “learned” features, in different prediction tasks relevant to organic and organometallic chemistry, where differing amounts of training data are available. These tasks include statistical models of stereo- and enantioselectivity, thermochemistry, and kinetics developed using experimental and quantum chemical data. The use of expert-guided molecular descriptors provides an opportunity to incorporate chemical knowledge, domain expertise, and physical constraints into statistical modeling. In applications to stereoselective organic and organometallic catalysis, where data sets may be relatively small and 3D-geometries and conformations play an important role, mechanistically informed features can be used successfully to obtain predictive statistical models that are also chemically interpretable. We provide an overview of several recent applications of this approach to obtain quantitative models for reactivity and selectivity, where topological descriptors, quantum mechanical calculations of electronic and steric properties, along with conformational ensembles, all feature as essential ingredients of the molecular representations used. Alternatively, more flexible, general-purpose molecular representations such as attributed molecular graphs can be used with machine learning approaches to learn the complex relationship between a structure and prediction target. This approach has the potential to out-perform more traditional representation methods such as “hand-crafted” molecular descriptors, particularly as data set sizes grow. One area where this is particularly relevant is in the use of large sets of quantum mechanical data to train quantitative structure–property relationships. A general approach toward curating useful data sets and training highly accurate graph neural network models is discussed in the context of organic bond dissociation enthalpies, where this strategy outperforms regression using precomputed descriptors. Finally, we describe how graph neural network predictions can be incorporated into mechanistically informed statistical models of chemical reactivity and selectivity. Once trained, this approach avoids the expensive computational overhead associated with quantum mechanical calculations, while maintaining chemical interpretability. We illustrate examples for which fast predictions of bond dissociation enthalpy and of the identities of radicals formed through cleavage of a molecule’s weakest bond are used in simple physical models of site-selectivity and reactivity.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
A Study of Complex Deep Learning Networks on High-Performance, Neuromorphic, and Quantum Computers
Current deep learning approaches have been very successful using convolutional neural networks trained on large graphical-processing-unit-based computers. Three limitations of this approach are that (1) they are based on a simple layered network topology, i.e., highly connected layers, without intra-layer connections; (2) the networks are manually configured to achieve optimal results, and (3) the implementation of the network model is expensive in both cost and power. In this paper, we evaluate deep learning models using three different computing architectures to address these problems: quantum computing to train complex topologies, high performance computing to automatically determine network topology, and neuromorphic computing for a low-power hardware implementation. We use the MNIST dataset for our experiment, due to input size limitations of current quantum computers. Our results show the feasibility of using the three architectures in tandem to address the above deep learning limitations. Finally, we show that a quantum computer can find high quality values of intra-layer connection weights in a tractable time as the complexity of the network increases, a high performance computer can find optimal layer-based topologies, and a neuromorphic computer can represent the complex topology and weights derived from the other architectures in low power memristive hardware.
97 MATHEMATICS AND COMPUTING↗
The effect of inhibition of PP1 and TNFα signaling on pathogenesis of SARS coronavirus
The complex interplay between viral replication and host immune response during infection remains poorly understood. While many viruses are known to employ antiimmune strategies to facilitate their replication, highly pathogenic virus infections can also cause an excessive immune response that exacerbates, rather than reduces pathogenicity. To investigate this dichotomy in severe acute respiratory syndrome coronavirus (SARS-CoV), we developed a transcriptional network model of SARS-CoV infection in mice and used the model to prioritize candidate regulatory targets for further investigation. We validated our predictions in 18 different knockout (KO) mouse strains, showing that network topology provides significant predictive power to identify genes that are important for viral infection. We identified a novel player in the immune response to virus infection, Kepi, an inhibitory subunit of the protein phosphatase 1 (PP1) complex, which protects against SARS-CoV pathogenesis. We also found that receptors for the proinflammatory cytokine, tumor necrosis factor alpha (TNFα), promote pathogenesis through a parallel feed-forward circuit that promotes inflammation. These results are consistent with previous studies showing the role of over-stimulation of the inflammatory response to SARS-CoV in pathogenesis. We conclude that the critical balance between immune response and inflammation can be manipulated to improve the outcome of the infection. Further, our study provides two potential therapeutic strategies for mitigating the effects of SARS-CoV infection, and may provide insight into treatment strategies for Middle East Respiratory Syndrome Coronavirus (MERS-CoV).
network inference↗
Structural Studies of Amorphous Materials by Fluctuation Electron Microscopy
Fluctuation Electron Microscopy (FEM) is a technique that examines the fluctuations in electron scattering across a uniformly thin amorphous sample. The statistics of the intensity fluctuations, mean and variance, reveal any underlying medium-range order present in the structure. The goals of this project were: (1) To determine the fundamentals of the scattering physics that gives rise to the variance signal in fluctuation electron microscopy (FEM); (2) To use these discoveries to find ways to quantify FEM; (3) To apply the FEM method to interesting and technologically important families of amorphous materials, particularly those with important applications in energy-related processes. Excellent progress was made in items (1) and (2). In stage (3) we did not examine the metamict zircons, as proposed. Instead, we examined films of polycrystalline and amorphous semi-conducting diamond. Significant accomplishments are: (1) A Reverse Monte Carlo procedure was successfully implemented to invert FEM data into a structural model. This is computer-intensive, but it demonstrated that diffraction and FEM data from amorphous silicon are most consistent with a paracrystallite model. This means that there is more diamond-like topology present in amorphous silicon than is predicted by the continuous random network model. (2) There is significant displacement decoherence arising in diffraction from amorphous silicon and carbon. The samples are being bombarded by the electron beam and atoms do not stay still while being irradiated – much more than was formerly understood. The atom motions cause the destructive and constructive interferences in the diffraction pattern to fluctuate with time, and it is the time-averaged speckle that is being measured. The variance is reduced by a factor m, 4 ≤ m ≤ 1000, relative to that predicted by kinematical scattering theory. (3) Speckle intensity obeys a gamma distribution, where the mean intensity $ \overline{I}\ $ and m are the two parameters governing the shape of the gamma distribution profile. m is determined by the illumination spatial coherence, which is normally very high, and mostly by the displacement decoherence within the sample. (4) Amorphous materials are more affected by the electron beam than are crystalline materials. Different samples exhibit different disruptibility, as measured by the effective values of m that fit the data. (5) Understanding the origin of the displacement decoherence better should lead to efficient methods for computing the observed variance from amorphous materials.
36 MATERIALS SCIENCE↗
Water Network Tool for Resilience (WNTR). User Manual, Version 0.2.3
The Water Network Tool for Resilience (WNTR, pronounced winter) is a Python package designed to simulate and analyze resilience of water distribution networks. Here, a network refers to the collection of pipes, pumps, valves, junctions, tanks, and reservoirs that make up a water distribution system. WNTR has an application programming interface (API) that is flexible and allows for changes to the network structure and operations, along with simulation of disruptive incidents and recovery actions. WNTR is based upon EPANET, which is a tool to simulate the movement and fate of drinking water constituents within distribution systems. Users are encouraged to be familiar with the use of EPANET and/or should have background knowledge in hydraulics and pressurized pipe network modeling before using WNTR. EPANET has a graphical user interface that might be a useful tool to facilitate the visualization of the network and the associated analysis results. Information on EPANET can be found at https://www.epa.gov/water-research/epanet. WNTR is compatible with EPANET 2.00.12 [Ross00]. In addition, users should have experience using Python, including the installation of additional Python packages. General information on Python can be found at https://www.python.org/.
54 ENVIRONMENTAL SCIENCES↗