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.
Oral presentation on NETL research delivered at the 2022 Resource Sustainability Meeting. This effort concerns advancing a concept for early kick detection through the development of synthetic (simulation and experiment) multiphase acoustic data corresponding to well characterized kick events. The simulation activities are the focus of the talk.
We present a deep machine learning algorithm to extract crystal field (CF) Stevens parameters from thermodynamic data of rare-earth magnetic materials. The algorithm employs a two-dimensional convolutional neural network (CNN) that is trained on magnetization, magnetic susceptibility and specific heat data that is calculated theoretically within the single-ion approximation and further processed using a standard wavelet transformation. We apply the method to crystal fields of cubic, hexagonal and tetragonal symmetry and for both integer and half-integer total angular momentum values J J of the ground state multiplet. We evaluate its performance on both theoretically generated synthetic and previously published experimental data on CeAgSb _2 2 , PrAgSb _2 2 and PrMg _2 2 Cu _9 9 , and find that it can reliably and accurately extract the CF parameters for all site symmetries and values of J J considered. This demonstrates that CNNs provide an unbiased approach to extracting CF parameters that avoids tedious multi-parameter fitting procedures.
Distributed scientific workflows are becoming more important with the interest in incorporating AI into their loops. A critical programming and performance question is how to compose workflow tasks when data is produced on one system but must be consumed on another. Since the dominant technique is composition with remote I/O, this paper explores its performance expectations. We describe BigFlowSim, a workflow I/O simulator that captures key implementation choices for remote I/O, including intensity, reuse, locality, access pattern, and data movement.With BigFlowSim, we generate a synthetic benchmark. We quantify the effects of each parameter with a performance sensitivity study. We explain trends in terms of data movement reduction and show that, under certain conditions, it is possible to establish a total order among most parameters. We apply these insights to a high energy physics workflow, Belle II Monte Carlo and simulate several I/O optimizations. Speedups range from 5% to 2×, without changing compute time.
We report a new deuterium–tritium experimental, DTE2, campaign has been conducted at the Joint European Torus (JET) between August 2021 and late December 2021. Motivated by significant enhancements in the past decade at JET, such as the ITER-like wall and enhanced auxiliary heating power, the campaign achieved a new fusion energy world record and performed a broad range of fundamental experiments to inform ITER physics scenarios and operations. New capabilities in the area of fusion product measurements by nuclear diagnostics were available as a result of a decade long enhancement program. These have been tested for the first time in DTE2 and a concise overview is provided here. Confined alpha particle measurements by gamma-ray spectroscopy were successfully demonstrated, albeit with limitations at neutron rates higher than some 10 17 n/s. High resolution neutron spectroscopy measurements with the magnetic proton recoil instrument were complemented by novel data from a set of synthetic diamond detectors, which enabled studies of the supra-thermal contributions to the neutron emission. In the area of escaping fast ion diagnostics, a lost fast ion detector and a set of Faraday cups made it possible to determine information on the velocity space and poloidal distribution of the lost alpha particles for the first time. This extensive set of data provides unique information for fundamental physics studies and validation of the numerical models, which are key to inform the physics and scenarios of ITER.
We present and validate a blinding method for the Lyman-$α$ (Ly$α$) forest analysis based on a modification of the Alcock-Paczynski test. In order to hide the background expansion history, the method employs a geometrical shift of each quasar (QSO) forest in wavelength space, once the quasar continuum has been fitted and the fluctuation field is extracted. The redshift positions for the QSO sample are also changed in a consistent manner. We show that the method remains effective when applied to real data, where contamination from metals and Lyman-$β$ is intrinsically mixed with the Lyman-$α$ forest. This limitation is primarily visible in the 1D correlation function, where other blinding strategies can mitigate the effect. To assess its effectiveness, the prescription is tested against a series of datasets of increasing complexity: from idealized low-noise mocks, to realistic DESI year one synthetic datasets, and finally to data from DESI first data release (DR1), using both the auto (Ly$α\times$Ly$α$) and cross (Ly$α\times$ QSO) correlations. We find that the method robustly shifts the BAO peak position from the 3D correlation functions to the expected value for cosmology changes of around 5% in the matter content, without altering the shape of the posteriors in the model parameters. In conclusion, this catalog-level blinding strategy is a viable method for cosmological inference with the Lyman-$α$ forest, particularly if a cross-analysis with other tracers using the same blinding strategy is pursued.
In this work, we consider the problem of low-rank approximation of massive dense nonnegative tensor data, for example, to discover latent patterns in video and imaging applications. As the size of data sets grows, single workstations are hitting bottlenecks in both computation time and available memory. We propose a distributed-memory parallel computing solution to handle massive data sets, loading the input data across the memories of multiple nodes, and performing efficient and scalable parallel algorithms to compute the low-rank approximation. We present a software package called Parallel Low-rank Approximation with Nonnegativity Constraints, which implements our solution and allows for extension in terms of data (dense or sparse, matrices or tensors of any order), algorithm (e.g., from multiplicative updating techniques to alternating direction method of multipliers), and architecture (we exploit GPUs to accelerate the computation in this work). We describe our parallel distributions and algorithms, which are careful to avoid unnecessary communication and computation, show how to extend the software to include new algorithms and/or constraints, and report efficiency and scalability results for both synthetic and real-world data sets.
Artificial Intelligence (AI) is revolutionizing physics research from probing the large-scale structure of the Universe to modeling subatomic interactions and fundamental forces. Yet, a major challenge persists: AI models trained on simulations or old experiment / astronomical survey often perform poorly when applied to new data exposing issues of dataset (domain) shift, model robustness, and uncertainty in predictions. This talk will introduce common challenges in applying AI across domains and present solutions based on domain adaptation a set of techniques designed to improve model generalization under domain shift. We will cover foundational ideas, practical strategies, and current research frontiers in this area. Through examples in astrophysics, we'll explore how domain adaptation can help bridge the gap between synthetic and real-world data, improve trust in model outputs, and advance scientific discovery.
Artificial Intelligence (AI) is revolutionizing physics research—from probing the large-scale structure of the Universe to modeling subatomic interactions and fundamental forces. Yet, a major challenge persists: AI models trained on simulations or old experiment / astronomical survey often perform poorly when applied to new data—exposing issues of dataset (domain) shift, model robustness, and uncertainty in predictions. This summer school session will introduce students to common challenges in applying AI across domains and present solutions based on domain adaptation—a set of techniques designed to improve model generalization under domain shift. We will cover foundational ideas, practical strategies, and current research frontiers in this area. Through examples in astrophysics, we'll explore how domain adaptation can help bridge the gap between synthetic and real-world data, improve trust in model outputs, and advance scientific discovery. The concepts discussed are broadly applicable across physics and other scientific disciplines, making this a valuable topic for anyone interested in building robust, transferable AI models for science.
In very-long baseline interferometric arrays, nearly co-located stations probe the largest scales and typically cannot resolve the observed source. In the absence of a large-scale structure, closure phases constructed with these stations are zero and, since they are independent of station-based errors, they can be used to probe data issues. Here, we show how these trivial closure phases become nonzero with a brightness distribution on smaller scales than their short baseline would suggest. When applied to sources that are made up of a bright compact and large-scale diffuse component, the trivial closure phases directly measure the centroid relative to the compact source and higher-order image moments. We present a technique to measure these image moments with minimal model assumptions and validate it on synthetic Event Horizon Telescope (EHT) data. We then apply this technique to 2017 and 2018 EHT observations of M87* and find a weak preference for extended emission in the direction of the large-scale jet. We also apply it to 2021 EHT data and measure the source centroid about 1 mas northwest of the compact ring, which is consistent with the jet observed at lower frequencies.
With the growing number of applications designed for heterogeneous HPC devices, application programmers and users are finding it challenging to compose scalable workflows as ensembles of these applications, that are portable, performant and resilient. The Kokkos C++ library has been designed to simplify this cumbersome procedure by providing an intra-application uniform programming model and portable performance. However, assembling multiple Kokkos-enabled applications into a complex workflow is still a challenge. Although Kokkos enables a uniform programming model, the inter-application data exchange still remains a challenge from both performance and software development cost perspectives. In order to address this issue, we propose Kokkos data staging memory space, an extension of Kokkos' data abstraction (memory space) for heterogeneous computing systems. This new abstraction allows to express data on a virtual shared-space for multiple Kokkos applications, thus extending Kokkos to support inter-application data exchange to build an efficient application workflow. Additionally, we study the effectiveness of asynchronous data layout conversions for applications requiring different memory access patterns for the shared data. Our preliminary evaluation with a synthetic benchmark indicate the effectiveness of this conversion adapted to three different scenarios representing access frequency and use patterns of the shared data.
We present a novel approach for modeling the loss factors of photovoltaic power generation systems (PV systems). This method is a white-box machine learning model built on convex optimization that is fast, interpretable, and auditable. It takes as an input the measured daily energy produced by the system, over a multi-year period, and returns a multiplicative decomposition model of the daily energy signal and full attribution of the total energy loss to each feature. The methods section of this paper has two major components: (1) the description of the signal decomposition (SD) model, expressed in the SD framework, and (2) the attribution of total energy losses via Shapley values. We validate the method on synthetic and open-source data sets and compare to similar methods from the literature.
Here, methods for estimating photoemission intensity from X-ray photoelectron spectroscopy data are examined. The role played by “synthetic” bell-shaped curves, integration intervals, background curves, and the use of relative sensitivity factors (RSFs) in reporting percentage atomic concentration for a sample is presented. In particular, photoemission lines with differing energy distributions obtained from the NaCl sample surface are used to demonstrate how a comparison of photoemission intensities is dependent on the line shapes, background curves, and appropriate use of RSFs.
Aboveground biomass (AGB) is a critical parameter to better understand the global carbon cycle and to develop sustainable forest management. However, a large uncertainty prevails. L-band SAR data have demonstrated strong potential to accurately retrieve AGB over low-biomass regions (<100 Mg ha-1). The upcoming NASA-ISRO Synthetic Aperture Radar mission will collect data at L- and S-band over earth’s landmass with a repeat period of 12 days, allowing us to have ample data for monitoring biomass and its dynamics. One of the key science requirements of the mission is to produce annual AGB maps at 1-ha resolution with RMS accuracy of 20 Mg/ha for 80 percentage of area over low-biomass regions in Calibration/Validation sites. The NISAR biomass algorithm will generate AGB maps based on the parameterization of semi-empirical model along with NISAR time-series dual pol data (HH and HV). To calibrate and validate the model for mission requirements, the mission will use reference estimates of AGB produced from ground inventory plots and airborne LiDAR data collected over selected sites distributed across different global ecoregions. This paper presents the initial results of the calibration/validation of the NISAR AGB retrieval algorithm over the Lenoir Landing (LENO), Alabama, USA site using NISAR simulated ALOS-2 time series data. Five multi-temporal dual-pol HH and HV NISAR Simulated ALOS 2 data collections were used as input to assess the performance of the model. The model AGB retrieval results shows that the NISAR model was able to achieve RMS accuracy within 20 Mg/ha.
The problem of noninstrusive load monitoring (NILM) is usually formulated as a single-channel blind source separation task, whose successful solution enable fast and convenient load identification and energy disaggregation. When applied at test time, NILM algorithms aim to identify the operating characteristics of individual appliances from an aggregate power measurement of the entire house. Recent advances in deep learning gave rise to many methods that mostly focus on learning a direct mapping from aggregate measurement to individual appliance power. However, these methods are not only computationally expensive, but they often suffer from overfitting and do not generalize very well. In this article, we propose a novel NILM method that leverages advances in statistical learning that have not been properly applied in this domain before. The proposed method consists of three stages: first, a Bayesian nonparametric learning-based approach for appliance state extraction; second, synthetic minority oversampling technique for data augmentation and mitigating the heavy imbalance in switching events; and third, appliance-specific lightweight long short-term memory networks for status classification for each appliance. Here, we adopt a “differential” input (the difference before and after the switching event) to reduce the complexity of network training and make the proposed method robust to multiappliance switching events. Experiments are conducted to demonstrate the effectiveness of the proposed method, achieving superior performance when compared to recent methods. An ablation study is conducted to demonstrate the effectiveness of each module of our method. Finally, we investigate the quality of generated synthetic samples.
Isosurface visualization is fundamental for exploring and analyzing 3D volumetric data. Marching cubes (MC) algorithms with linear interpolation are commonly used for isosurface extraction and visualization. Although linear interpolation is easy to implement, it has limitations when the underlying data is complex and high-order, which is the case for most real-world data. Linear interpolation can output vertices at the wrong location. Its inability to deal with sharp features and features smaller than grid cells can lead to an incorrect isosurface with holes and broken pieces. Despite these limitations, isosurface visualizations typically do not include insight into the spatial location and the magnitude of these errors. We utilize high-order interpolation methods with MC algorithms and interactive visualization to highlight these uncertainties. Our visualization tool helps identify the regions of high interpolation errors. It also allows users to query local areas for details and compare the differences between isosurfaces from different interpolation methods. In addition, we employ high-order methods to identify and reconstruct possible features that linear methods cannot detect. We showcase how our visualization tool helps explore and understand the extracted isosurface errors through synthetic and real-world data.
ABSTRACT Motivation Biomedical machine reading comprehension (biomedical-MRC) aims to comprehend complex biomedical narratives and assist healthcare professionals in retrieving information from them. The high performance of modern neural network-based MRC systems depends on high-quality, large-scale, human-annotated training datasets. In the biomedical domain, a crucial challenge in creating such datasets is the requirement for domain knowledge, inducing the scarcity of labeled data and the need for transfer learning from the labeled general-purpose (source) domain to the biomedical (target) domain. However, there is a discrepancy in marginal distributions between the general-purpose and biomedical domains due to the variances in topics. Therefore, direct-transferring of learned representations from a model trained on a general-purpose domain to the biomedical domain can hurt the model’s performance. Results We present an adversarial learning-based domain adaptation framework for the biomedical machine reading comprehension task (BioADAPT-MRC), a neural network-based method to address the discrepancies in the marginal distributions between the general and biomedical domain datasets. BioADAPT-MRC relaxes the need for generating pseudo labels for training a well-performing biomedical-MRC model. We extensively evaluate the performance of BioADAPT-MRC by comparing it with the best existing methods on three widely used benchmark biomedical-MRC datasets—BioASQ-7b, BioASQ-8b and BioASQ-9b. Our results suggest that without using any synthetic or human-annotated data from the biomedical domain, BioADAPT-MRC can achieve state-of-the-art performance on these datasets. Availability and implementation BioADAPT-MRC is freely available as an open-source project at https://github.com/mmahbub/BioADAPT-MRC. Supplementary information Supplementary data are available at Bioinformatics online.
The Mapper algorithm is a visualization technique in topological data analysis (TDA) that outputs a graph reflecting the structure of a given dataset. However, the Mapper algorithm requires tuning several parameters in order to generate a “nice” Mapper graph. This paper focuses on selecting the cover parameter. We present an algorithm that optimizes the cover of a Mapper graph by splitting a cover repeatedly according to a statistical test for normality. Our algorithm is based on G-means clustering, which searches for the optimal number of clusters in 𝑘-means by iteratively applying the Anderson–Darling test. Our splitting procedure employs a Gaussian mixture model to carefully choose the cover according to the distribution of the given data. In conclusion, experiments for synthetic and real-world datasets demonstrate that our algorithm generates covers so that the Mapper graphs retain the essence of the datasets, while also running significantly faster than a previous iterative method.
Microbial communities are diverse biological systems that include taxa from across multiple kingdoms of life. Notably, interactions between bacteria and fungi play a significant role in determining community structure. However, these statistical associations across kingdoms are more difficult to infer than intra-kingdom associations due to the nature of the data involved using standard network inference techniques. We quantify the challenges of cross-kingdom network inference from both theoretical and practical points of view using synthetic and real-world microbiome data. We detail the theoretical issue presented by combining compositional data sets drawn from the same environment, e.g. 16S and ITS sequencing of a single set of samples, and we survey common network inference techniques for their ability to handle this error. We then test these techniques for the accuracy and usefulness of their intra- and inter-kingdom associations by inferring networks from a set of simulated samples for which a ground-truth set of associations is known. We show that while the two methods mitigate the error of cross-kingdom inference, there is little difference between techniques for key practical applications including identification of strong correlations and identification of possible keystone taxa (i.e. hub nodes in the network). Furthermore, we identify a signature of the error caused by transkingdom network inference and demonstrate that it appears in networks constructed using real-world environmental microbiome data.