Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “time series encoding”

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.

41 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Bundle Networks: Fiber Bundles, Local Trivializations, and a Generative Approach to Exploring Many-to-one Maps

Many-to-one maps are ubiquitous in machine learning, from the image recognition model that assigns a multitude of distinct images to the concept of ``cat'' to the time series forecasting model which assigns a range of distinct time-series to a single scalar regression value. While the primary use of such models is naturally to associate correct output to each input, in many problems it is also useful to be able to explore, understand, and sample from a model's fibers, which are the set of input values x such that f(x) = y, for fixed y in the output space. In this paper we show that popular generative architectures are ill-suited to such tasks. Motivated by this we introduce a novel generative architecture, a Bundle Network, based on the concept of a fiber bundle from (differential) topology. BundleNets exploit the idea of a local trivialization wherein a space can be locally decomposed into a product space that cleanly encodes the many-to-one nature of the map. By enforcing this decomposition in BundleNets and by utilizing state-of-the-art invertible components, investigating a network's fibers becomes natural.

Courts, Nicolas C.↗
DOE OSTI2022

Deep learning to estimate permeability using geophysical data

Time-lapse electrical resistivity tomography (ERT) is a popular geophysical method to estimate three-dimensional (3D) permeability fields from electrical potential difference measurements. Traditional inversion and data assimilation methods are used to ingest this ERT data into hydrogeophysical models to estimate permeability. Due to ill-posedness and the curse of dimensionality, existing inversion strategies provide poor estimates and low resolution of the 3D permeability field. Recent advances in deep learning provide us with powerful algorithms to overcome this challenge. This paper presents a deep learning (DL) framework to estimate the 3D subsurface permeability from time-lapse ERT data. To test the feasibility of the proposed framework, we train DL-enabled inverse models on simulation data. Each measurement in both synthetic and field data is standardized by removing the mean and scaling the time-series to unit variance. This pre-processing step is necessary to bring simulation data closer to field observations. Subsurface process models based on hydrogeophysics are used to generate this synthetic data. Training performed on limited simulation data resulted in the DL model over-fitting. An advanced data augmentation based on mixup is implemented to generate additional training samples to overcome this issue. This mixup technique creates weakly labeled (low-fidelity) samples from strongly labeled (high-fidelity) data. The weakly labeled training data is then used to develop DL-enabled inverse models and reduce over-fitting. As both time-lapse ERT (1133048 features/realization) and 3D permeability (585453 features/realization) data samples are from a high-dimensional space, principal component analysis (PCA) is employed to reduce dimensionality. Encoded ERT and encoded permeability are generated using the trained PCA estimators. A deep neural network is then trained to map the encoded ERT to encoded permeability. This mixup training and unsupervised learning allowed us to build a fast and reasonably accurate DL-based inverse model under limited simulation data. Results show that proposed weak supervised learning can capture salient spatial features in the 3D permeability field. Quantitatively, the average mean squared error (in terms of the natural log) on the strongly labeled training, validation, and test datasets is less than 0.5. The R 2 -score (global metric) is greater than 0.75, and the percent error in each cell (local metric) is less than 10%. Finally, an added benefit in terms of computational cost is that the proposed DL-based inverse model is at least O(10 4 ) times faster than running a forward model once it is trained. Data generation, DL model training, and hyperparameter tuning to identify optimal neural network architectures utilized high-performance computing resources while the DL inference is performed on a standard laptop. Approximately, O(10 5 ) processor hours are used for generating data and DL tuning and training. We acknowledge that the data generation and DL model development are expensive. But once a DL model is trained, it can be re-used for inversion rapidly for the given system, with set physics and domain. Note that traditional inversion may require multiple forward model simulations (e.g., in the order of 10 to 1000), which are very expensive. This computational savings ≈ O(10 5 ) – O(10 7 )) makes the proposed DL-based inverse model attractive for subsurface imaging and real-time ERT monitoring applications due to fast and yet reasonably accurate estimations of permeability field.

58 GEOSCIENCES↗
DOE OSTI2020

The cosmological bootstrap: inflationary correlators from symmetries and singularities

Scattering amplitudes at weak coupling are highly constrained by Lorentz invariance, locality and unitarity, and depend on model details only through coupling constants and the particle content of the theory. For example, four-particle amplitudes are analytic for contact interactions and have simple poles with appropriately positive residues for tree-level exchange. In this paper, we develop an understanding of inflationary correlators which parallels that of flat-space scattering amplitudes. Specifically, we study slow-roll inflation with weak couplings to extra massive particles, for which all correlation functions are controlled by an approximate conformal symmetry on the boundary of the spacetime. After systematically classifying all possible contact terms in de Sitter space, we derive an analytic expression for the four-point function of conformally coupled scalars mediated by the tree-level exchange of massive scalars. Conformal symmetry implies that the correlator satisfies a pair of differential equations with respect to spatial momenta, encoding bulk time evolution in purely boundary terms. The absence of unphysical singularities (and the correct normalization of physical ones) completely fixes this correlator. Moreover, a “spin-raising” operator relates it to the correlators associated with the exchange of particles with spin, while “weight-shifting” operators map it to the four-point function of massless scalars. We explain how these de Sitter four-point functions can be perturbed to obtain inflationary three-point functions. Using our formalism, we reproduce many classic results in the literature, such as the three-point function of slow-roll inflation, and provide a complete classification of all inflationary three- and four-point functions arising from weakly broken conformal symmetry. Remarkably, the inflationary bispectrum associated with the exchange of particles with arbitrary spin is completely characterized by the soft limit of the simplest scalar-exchange four-point function of conformally coupled scalars and a series of contact terms. Finally, we demonstrate that the inflationary correlators contain flat-space scattering amplitudes via a suitable analytic continuation of the external momenta, which can also be directly connected with the signals for particle production seen in the squeezed limit.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

The Swan: Data-driven Inference of Stellar Surface Gravities for Cool Stars from Photometric Light Curves

Stellar light curves are well known to encode physical stellar properties. Precise, automated, and computationally inexpensive methods to derive physical parameters from light curves are needed to cope with the large influx of these data from space-based missions such as Kepler and TESS. Here we present a new methodology that we call “The Swan,” a fast, generalizable, and effective approach for deriving stellar surface gravity (logg) for main-sequence, subgiant, and red giant stars from Kepler light curves using local linear regression on the full frequency content of Kepler long-cadence power spectra. With this inexpensive data-driven approach, we recover logg to a precision of ~0.02 dex for 13,822 stars with seismic logg values between 0.2 and 4.4 dex and ~0.11 dex for 4646 stars with Gaia-derived logg values between 2.3 and 4.6 dex. We further develop a signal-to-noise metric and find that granulation is difficult to detect in many cool main-sequence stars (T {sub eff} ≲ 5500 K), in particular K dwarfs. By combining our logg measurements with Gaia radii, we derive empirical masses for 4646 subgiant and main-sequence stars with a median precision of ~7%. Finally, we demonstrate that our method can be used to recover logg to a similar mean absolute deviation precision for a TESS baseline of 27 days. Our methodology can be readily applied to photometric time series observations to infer stellar surface gravities to high precision across evolutionary states.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2026

Predicting river turbidity in Pine Island Bayou using machine learning techniques coupled with variational mode decomposition

Elevated turbidity levels pose significant public health risks by facilitating the transport of harmful pollutants, including metals, organic compounds, and pathogenic microorganisms into the surface water. These conditions create serious challenges for public recreational water use and drinking water treatment, leading to economic losses and health risks. This study utilizes water monitoring data in Pine Island Bayou, Texas, and develops a Sequence-to-Sequence (S2S) model to predict turbidity using Attention-based Gated Recurrent Units with Encoder-Decoder (AT-GRU-ED) and Long Short-Term Memory (LSTM), coupled with Variational Mode Decomposition (VMD). Compared to the model without VMD, the model demonstrates satisfactory 72-hour turbidity prediction performance, achieving MAEs of 2.60 and 3.29 NTU (reductions of 53% and 58%), RMSEs of 21.08 and 31.49 NTU (reductions of 82% and 80%), and R² values of 0.96 and 0.84 on the validation and test sets, respectively. Feature importance analysis reveals that water temperature is the dominant factor influencing seasonal turbidity patterns, while real-time hourly rainfall significantly contributes to short-term variability. Turbidity typically peaks within 48 hours after rainfall events due to lagged effects from surface runoff and upstream flow. Findings suggest suspending recreational water use and water supply pumping for three days after heavy rainfall can benefit public health and improve water treatment processes. Discharges above 100 m3/s are found to accelerate sediment dilution and transport, reducing turbidity levels more quickly after the peak. In conclusion, the proposed model demonstrates reliable 72-hour turbidity prediction, supporting decision-making for water treatment plant operations and providing early warning for public recreational water use.

Deep learning↗