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.
R-stream release 3.22.3 including the Glow and TVM front-ends, and the features developed in the Portacy 2B, X-MAT and STAPL projects. R-Stream is an automatic parallelization, source-to-source compiler. The TVM and Glow front-ends to R-Stream were developed to automatically map neural net models to parallel computers using openMP and CUDA (for Glow).
The QC-GN2OMS2 neural network was designed to predict mass spectra better than currently available approaches. Fundamentally the neural net is designed to input the structure of a molecule and collision energy value then output a spectra datafile for the end user. This software is able to make predictions orders of magnitude faster (milliseconds) than QC methods and with 50-60% greater accuracy than rules approaches.
Data-based closure models are increasingly being used to replace physic-based closure models because of their flexibility and the growing availability of data. However, closure models are subject to uncertainty because of lack of data in parts of the input space (epistemic uncertainty) or because of noise in the data (aleatoric uncertainty). This software contains a toolbox for training bayesian neural nets that can estimate both uncertainties. A specific treatment is provided to ensure accuracy outside of the data distribution. A set of tools are provided to reduce the dimensionality of the uncertain parameter space, thereby enabling fast uncertainty propagation.
SAND2024-13788O pyKinML software is used to train neural net potential energy surfaces for hydrocarbon molecules. The models study reaction kinetics but can be used for other purposes as well, such as running molecular dynamics simulation. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.
The CHM-MS-net Canopy Height Model is a MS-net neural network applied to high resolution satellite imagery to estimate tree height, tree location, and crown radius.
We developed a computational strategy to correlate bulk combustion metrics of novel fuels and blends in the low-temperature autoignition regime with measurements of key combustion intermediates in a small-volume, dilute, high-pressure reactor. We used neural net analysis of a large simulation dataset to obtain an approximate correlation and proposed experimental and computational steps needed to refine such a predictive correlation. We also designed and constructed a high-pressure laboratory apparatus to conduct the proposed measurements and demonstrated its performance on three canonical fuels: n-heptane, i-octane, and dimethyl ether.
Photon- and field-induced charge transfer has central importance in the generation and storage of electricity, the novel properties of materials, photo-induced catalysis, and electro-optic activity (e.g., photovoltaic cells, fuel cells, and organic chromophores for use in optical fibers and light-emission diodes). These non-equilibrium electronic and chemical transformations are probed by ultrafast, nonlinear spectroscopies. Accurate simulations play a crucial role in our ability to understand, optimize, and control these transformations. This project applies modern deep learning and machine learning (ML) methods to dramatically improve models of electronic dynamics, electronic-nuclear dynamics, and spectroscopic measurements for improved simulations of chemistry in complex environments, far from equilibrium phenomena, and processes in extreme environments, such as materials exposed to strong or resonant fields. This project develops accurate neural net models that go beyond predictive capability to also provide new insight into the fundamental physics underlying electron and nuclear dynamics. To achieve its objectives, this project explores and develops customized versions of high-capacity deep learning algorithms/models. These techniques are developed with an emphasis on fundamental chemical insight, not just predictive accuracy, to assist the development of the next generation of quantum simulation methods.
Computational Science is being revolutionized by integrating AI and simulation and, in particular, by deep learning surrogate models that can replace all or part of traditional large‐scale HPC computations. Such surrogates can achieve remarkable performance improvements, as much as several orders of magnitude, and save both compute time and energy. The Surrogate Benchmark Initiative (SBI) project creates a community repository and FAIR (Findable, Accessible, Interoperable, and Reusable) data ecosystem for HPC application surrogate benchmarks. The SBI team comes from Argonne National Laboratory (ANL), Indiana University (IU), Rutgers University, the University of Tennessee, Knoxville (UTK), and the University of Virginia(UVA). SBI repositories include data, code, and all relevant collateral artifacts, that the science and engineering community needs to use and reuse these data sets and surrogates. SBI repositories generate active research from both participants in SBI and the broader AI and domain science communities. This project develops surrogates that use several different neural nets to learn and quickly infer the results of simulations and data systems and capture them as surrogate benchmarks with a rich set of metadata, covering. Data; Model; Metrics specification; Machine specification; Science, Speed, Power Results, We research FAIR metadata for these benchmarks. We develop application surrogate examples as benchmarks across many fields (ANL, UTK, IU, UVA). We also study non Surrogate benchmarks that have many common features and similar issues regarding FAIRness. We work with MLCommons (UVA, UTK), which is a major machine learning benchmarking activity where we get metadata ontologies, software, and benchmarks, benchmarks have datasets, models, and metadata, and they need a technical framework developed by UTK and Rutgers and deployed by UVA. We study features of Surrogates, including performance, training set size, and uncertainty quantification (Rutgers, UVA and IU).
Computational Science is being revolutionized by integrating AI and simulation and, in particular, by deep learning surrogate models that can replace all or part of traditional large‐scale HPC computations. Such surrogates can achieve remarkable performance improvements, as much as several orders of magnitude, and save both compute time and energy. The Surrogate Benchmark Initiative (SBI) project creates a community repository and FAIR (Findable, Accessible, Interoperable, and Reusable) data ecosystem for HPC application surrogate benchmarks. The SBI team comes from Argonne National Laboratory (ANL), Indiana University (IU), Rutgers University, the University of Tennessee, Knoxville (UTK), and the University of Virginia (UVA). SBI repositories include data, code, and all relevant collateral artifacts that the science and engineering community need to use and reuse these data sets and surrogates. SBI repositories generate active research from both the participants in SBI and the broad community of AI and domain scientists. This project develops surrogates that use several different neural nets to learn and quickly infer the results of simulations and data systems and captures them as surrogate benchmarks with a rich set of metadata covering: Data; Model; Metrics specification; Machine specification; and Science, Speed, and Power Results. We research FAIR metadata for these benchmarks. We develop application surrogate examples as benchmarks across many fields (ANL, UTK, IU, UVA). We also study non-Surrogate benchmarks that have many common features and similar issues as regards FAIRness. We work with MLCommons (UVA, UTK), which is a major machine learning benchmarking activity where we get metadata ontologies, software, and benchmarks, Benchmarks have datasets, models, and metadata and they need a technical framework developed by UTK and Rutgers and deployed by UVA. We study features of Surrogates including performance, training set size, and uncertainty quantification (Rutgers, UVA and IU).
Abstract The Active Optics System of the Vera C. Rubin Observatory (Rubin) uses information provided by four wave front sensors to determine deviations between the reconstructed wave front and the ideal wave front. The observed deviations are used to adjust the control parameters of the optical system to maintain image quality across the 3.°5 field of view. The baseline approach from the project is to obtain amplitudes of the Zernike polynomials describing the distorted wave front from out-of-focus images collected by the wave front sensors. These Zernike amplitudes are related via an “influence matrix” to the control parameters necessary to correct the wave front. In this paper, we use deep-learning methods to extract the control parameters directly from the images captured by the wave front sensors. Our neural net model uses anti-aliasing pooling to boost performance, and a domain-specific loss function to aid learning and generalization. The accuracy of the control parameters derived from our model exceeds Rubin requirements even in the presence of full-moon background levels and mis-centering of reference stars. Although the training process is time consuming, model evaluation requires only a few milliseconds. This low latency should allow for the correction of the optical configuration during the readout and slew interval between successive exposures.
The application of artificial intelligence to spacecraft and aerospace systems is discussed. Expert systems, robotics, space station automation, fault diagnostics, parallel processing, knowledge representation, scheduling, man-machine interfaces and neural nets are among the topics discussed.
Transmitting large amounts of data efficiently among neighboring stars will vitally support any eventual contact with extrasolar intelligence, whether alien or human. Laser carriers are particularly suitable for high-quality, targeted links. Space laser transmitter systems designed by this work, based on both demonstrated and imminent advanced space technology, could achieve reliable data transfer rates as high as 1 kb/s to matched receivers as far away as 25 pc, a distance including over 700 approximately solar-type stars. The centerpiece of this demonstration study is a fleet of automated spacecraft incorporating adaptive neural-net optical processing active structures, nuclear electric power plants, annular momentum control devices, and ion propulsion. Together the craft sustain, condition, modulate, and direct to stellar targets an infrared laser beam extracted from the natural mesospheric, solar-pumped, stimulated CO2 emission recently discovered at Venus. For a culture already supported by mature interplanetary industry, the cost of building planetary or high-power space laser systems for interstellar communication would be marginal, making such projects relevant for the next human century. Links using high-power lasers might support data transfer rates as high as optical frequencies could ever allow. A nanotechnological society such as we might become would inevitably use 10 to the 20th power b/yr transmission to promote its own evolutionary expansion out of the galaxy.
Topics relative to the application of artificial intelligence to space operations are discussed. New technologies for space station automation, design data capture, computer vision, neural nets, automatic programming, and real time applications are discussed.
The Hubble Space Telescope (HST) presents an especially challenging scheduling problem since a year's observing program encompasses tens of thousands of exposures facing numerous coupled constraints. Recent progress in the development of planning and scheduling tools is discussed which augment the existing HST ground system. General methods for representing activities, constraints, and constraint satisfaction, and time segmentation were implemented in a scheduling testbed. The testbed permits planners to evaluate optimal scheduling time intervals, calculate resource usage, and to generate long and medium range plans. Graphical displays of activities, constraints, and plans are an important feature of the system. High-level scheduling strategies using rule based and neural net approaches were implemented.
The size of the basins of attraction around fixed points in recurrent neural nets (NNs) can be modified by a training process. Controlling these attractive regions by presenting training data with various amount of noise added to the prototype signal vectors is discussed. Application of this technique to signal processing results in a classification system whose sensitivity can be controlled. This new technique is applied to the classification of temporal sequences in telemetry data.
The main objective of this research is to develop an algorithm for isolated-word recognition. This research is focused on digital signal analysis rather than linguistic analysis of speech. Features extraction is carried out by applying a Linear Predictive Coding (LPC) algorithm with order of 10. Continuous-word and speaker independent recognition will be considered in future study after accomplishing this isolated word research. To examine the similarity between the reference and the training sets, two approaches are explored. The first is implementing traditional pattern recognition techniques where a dynamic time warping algorithm is applied to align the two sets and calculate the probability of matching by measuring the Euclidean distance between the two sets. The second is implementing a backpropagation artificial neural net model with three layers as the pattern classifier. The adaptation rule implemented in this network is the generalized least mean square (LMS) rule. The first approach has been accomplished. A vocabulary of 50 words was selected and tested. The accuracy of the algorithm was found to be around 85 percent. The second approach is in progress at the present time.
The ability of feed-forward neural net architectures to learn continuous-valued mappings in the presence of noise is demonstrated in relation to parameter identification and real-time adaptive control applications. Factors and parameters influencing the learning performance of such nets in the presence of noise are identified. Their effects are discussed through a computer simulation of the Back-Error-Propagation algorithm by taking the example of the cart-pole system controlled by a nonlinear control law. Adequate sampling of the state space is found to be essential for canceling the effect of the statistical fluctuations and allowing learning to take place.
It is proposed that chaotic attractors incorporated in neural net models can represent classes of patterns in the same way in which a set of static attractors represent unrelated patterns. Therefore, chaotic states of neuron activity are associated with higher level cognitive processes such as generalization and abstraction.