Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data inference”

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.

At least 631 records · Page 35
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Fourier-DeepONet: Fourier-enhanced deep operator networks for full waveform inversion with improved accuracy, generalizability, and robustness

In this article, full waveform inversion (FWI) infers the subsurface structure information from seismic waveform data by solving a non-convex optimization problem. Data-driven FWI has been increasingly studied with various neural network architectures to improve accuracy and computational efficiency. Nevertheless, the applicability of pre-trained neural networks is severely restricted by potential discrepancies between the source function used in the field survey and the one utilized during training. Here, we develop a Fourier-enhanced deep operator network (Fourier-DeepONet) for FWI with the generalization of seismic sources, including the frequencies and locations of sources. Specifically, we employ the Fourier neural operator as the decoder of DeepONet, and we utilize source parameters as one input of Fourier-DeepONet, facilitating the resolution of FWI with variable sources. To test Fourier-DeepONet, we develop three new and realistic FWI benchmark datasets (FWI-F, FWI-L, and FWI-FL) with varying source frequencies, locations, or both. Our experiments demonstrate that compared with existing data-driven FWI methods, Fourier-DeepONet obtains more accurate predictions of subsurface structures in a wide range of source parameters. Moreover, the proposed Fourier-DeepONet exhibits superior robustness when handling data with Gaussian noise or missing traces and sources with Gaussian noise, paving the way for more reliable and accurate subsurface imaging across diverse real conditions.

42 ENGINEERING↗
DOE OSTI2021

Bayesian calibration of strength model parameters from Taylor impact data

Materials strength plays a key role in determining the mechanical response of engineered structures. As such, accurate strength models are crucial in simulations involving complex loading conditions, particularly when deformation in the plastic regime is deemed important. In this work, a Gaussian process based surrogate for a finite element simulation of a Taylor impact test is developed and used for Bayesian calibration of the Preston–Tonks–Wallace strength model. Here, the surrogate model is shown to closely approximate the salient features of the Taylor cylinder deformation and is validated against simulation output before being used in the strength model calibration routine. The results show that Taylor impact test data can be used in the calibration of constitutive equations through the use of a combination of data science techniques, namely Gaussian processes and Bayesian inference.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Uranium sorption to organic matter and long-term accumulation in a pristine alpine wetland

Understanding the controls on uranium (U) mobility in the environment is key to improve the management of sites contaminated by U mining activities. Previous research has shown that natural or engineered wetlands are particularly able to scavenge high amounts of U(VI) and U(IV) under noncrystalline forms. However, questions remain on the respective roles of sorption and reduction processes in the removal of U from running waters in wetlands, as well as on the long-term stability of U storage. Here, we performed a series of geochemical, isotopic ($δ$ 238 U, ( 234 U/ 238 U)), microscopic (SEM-EDXS, EPMA) and spectroscopic (µ-XRF, µ-XAS, XANES and EXAFS at the U L 3 and M 4 -edges and Fe K-edge) investigations to determine the modes of U accumulation and assess U mobility in a natural exceptionally U-enriched (up to 5000 µg/g) wetland on the shore of Lake Nègre (Mediterranean Alps, France). Uranium (VI) was largely dominant in the two studied soil cores, except a few samples containing as much as ~50 % U(IV). At the particle scale, U is associated to a variety of organic constituents of the soil matrix with a homogenous oxidation state. Bulk EXAFS spectroscopy at the U L 3 -edge shows that U is mostly mononuclear, with dominant monodentate binding to organic moieties (C neighbors at ~3.45 Å). An additional minor fraction of U under polymeric forms is inferred from wavelet (CCWT) analysis of the EXAFS data. These observations are reinforced by 1 M bicarbonate extractions that result in the dissolution of 82–96 % of total U, including putative polymeric species. At the wetland scale, similar or slightly fractionated isotopic ratios ($δ$ 238 U) between the wetland-feeding creek waters and the wetland soils are observed, supporting the idea that U(VI) sorption on organic matter is the primary U scavenging mechanism. Furthermore, it indicates that partial U(VI) reduction to U(IV) occurs as a second step, after sorption. Analysis of U decay chain disequilibria in the cores as a function of depth suggests that U accumulation in this wetland has lasted for several thousand years. In conclusion, we propose that the wetland acts as an active reactor where U has been massively accumulating for ~14500 years, especially as U(VI) forms associated to organic matter and is further partly exported to the lake through soil erosion.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗
DOE OSTI2021

Genetic program activity delineates risk, relapse, and therapy responsiveness in multiple myeloma

Despite recent advancements in the treatment of multiple myeloma (MM), nearly all patients ultimately relapse and many become refractory to multiple lines of therapies. Therefore, we not only need the ability to predict which patients are at high risk for disease progression but also a means to understand the mechanisms underlying their risk. Here, we report a transcriptional regulatory network (TRN) for MM inferred from cross-sectional multi-omics data from 881 patients that predicts how 124 chromosomal abnormalities and somatic mutations causally perturb 392 transcription regulators of 8549 genes to manifest in distinct clinical phenotypes and outcomes. We identified 141 genetic programs whose activity profiles stratify patients into 25 distinct transcriptional states and proved to be more predictive of outcomes than did mutations. The coherence of these programs and accuracy of our network-based risk prediction was validated in two independent datasets. We observed subtype-specific vulnerabilities to interventions with existing drugs and revealed plausible mechanisms for relapse, including the establishment of an immunosuppressive microenvironment. Investigation of the t(4;14) clinical subtype using the TRN revealed that 16% of these patients exhibit an extreme-risk combination of genetic programs (median progression-free survival of 5 months) that create a distinct phenotype with targetable genes and pathways.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Preventing Reverse Engineering of Critical Industrial Data with DIOD

Business analytics augmented by artificial intelligence and machine learning (AI/ML) have revolutionized the role of data in the modern world. In recent years, businesses have incorporated data into their decision-making process for better prediction, risk-assessment, content creation, etc. While such businesses often seek to leverage the full use of their data through third-party AI/ML services, they are often hampered by the risks of data leaks, reverse-engineering, stolen technology, etc. that often have disastrous consequences for businesses and their stakeholders alike. Thus, there arises a need for data masking prior to its transmission that obfuscates proprietary information while preserving the information relevant for AI/ML applications. In order to meet the needs of industrial data which are significantly different from those of data warehouses, previous work proposed an efficient time and space-scalable data masking paradigm known as the deceptive infusion of data (DIOD) methodology. The present work expands upon this work by leveraging existing reverse-engineering capabilities to facilitate the decomposition of industrial data into its proprietary and AI/ML-relevant parts, referred to as fundamental and inference metadata respectively. Both sets of metadata are further obfuscated in accordance with the DIOD methodology to create the DIOD rendition of the industrial data, which is rendered immune to reverse-engineering by discarding proprietary information and only preserving AI/ML-relevant information. Additionally, constraints of the original DIOD manuscript are relaxed using mutual information by configuring the methodology to the target AI/ML application to unlock the full potential of the DIOD methodology. As an example, data from a nuclear reactor is transformed into that from a nonlinear spring-mass system with different levels of data masking as required by the generic system and the target application.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Particle image velocimetry analysis with simultaneous uncertainty quantification using Bayesian neural networks

Particle image velocimetry (PIV) is an effective tool in experimental fluid mechanics for extracting flow fields from images. Recently, convolutional neural networks (CNNs) have been used to perform PIV analysis with accuracy on par with classical methods. Here we extend the use of CNNs to analyze PIV data while providing simultaneous uncertainty quantification on the inferred flow field. The method we apply in this paper is a Bayesian convolutional neural network (BCNN) which learns distributions of the CNN weights through variational Bayes. In order to demonstrate the utility of BCNNs for the PIV task, we compare the performance of three distinct BCNN models with simple architectures. The first network estimates flow velocity from image interrogation regions only. Our second model learns to infer velocity from both the image interrogation regions and interrogation region cross-correlation maps. Finally, our best performing network infers velocities from interrogation region cross-correlation maps only. We find that BCNNs using interrogation region cross-correlation maps as inputs perform better than those using interrogation windows only as inputs and discuss reasons why this may be the case. Additionally, we test the best performing BCNN on a full synthetic test image pair and a real image pair from the 1st International PIV Challenge. We show that ~98% of true particle displacements from the full synthetic image pair can be captured within the BCNN's 95% confidence intervals, and that the BCNN's performance on the real image pair is quantitatively similar to that of algorithms tested in the 1st International PIV Challenge. Finally, we show that BCNNs can be generalized to be used with multi-pass PIV algorithms with a moderate loss in accuracy, which may be overcome by future work on finetuning and training schemes. So to our knowledge, this is the first use of Bayesian neural networks to perform PIV.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

Deep probabilistic direction prediction in 3D with applications to directional dark matter detectors

Abstract We present the first method to probabilistically predict 3D direction in a deep neural network model. The probabilistic predictions are modeled as a heteroscedastic von Mises-Fisher distribution on the sphere S 2 , giving a simple way to quantify aleatoric uncertainty. This approach generalizes the cosine distance loss which is a special case of our loss function when the uncertainty is assumed to be uniform across samples. We develop approximations required to make the likelihood function and gradient calculations stable. The method is applied to the task of predicting the 3D directions of electrons, the most complex signal in a class of experimental particle physics detectors designed to demonstrate the particle nature of dark matter and study solar neutrinos. Using simulated Monte Carlo data, the initial direction of recoiling electrons is inferred from their tortuous trajectories, as captured by the 3D detectors. For 40 keV electrons in a 70% He 30% CO 2 gas mixture at STP, the new approach achieves a mean cosine distance of 0.104 (26 ∘ ) compared to 0.556 (64 ∘ ) achieved by a non-machine learning algorithm. We show that the model is well-calibrated and accuracy can be increased further by removing samples with high predicted uncertainty. This advancement in probabilistic 3D directional learning could increase the sensitivity of directional dark matter detectors.

Computer Science↗
DOE OSTI2021

Deciphering enhancer sequence using thermodynamics-based models and convolutional neural networks

Abstract Deciphering the sequence-function relationship encoded in enhancers holds the key to interpreting non-coding variants and understanding mechanisms of transcriptomic variation. Several quantitative models exist for predicting enhancer function and underlying mechanisms; however, there has been no systematic comparison of these models characterizing their relative strengths and shortcomings. Here, we interrogated a rich data set of neuroectodermal enhancers in Drosophila, representing cis- and trans- sources of expression variation, with a suite of biophysical and machine learning models. We performed rigorous comparisons of thermodynamics-based models implementing different mechanisms of activation, repression and cooperativity. Moreover, we developed a convolutional neural network (CNN) model, called CoNSEPT, that learns enhancer ‘grammar’ in an unbiased manner. CoNSEPT is the first general-purpose CNN tool for predicting enhancer function in varying conditions, such as different cell types and experimental conditions, and we show that such complex models can suggest interpretable mechanisms. We found model-based evidence for mechanisms previously established for the studied system, including cooperative activation and short-range repression. The data also favored one hypothesized activation mechanism over another and suggested an intriguing role for a direct, distance-independent repression mechanism. Our modeling shows that while fundamentally different models can yield similar fits to data, they vary in their utility for mechanistic inference. CoNSEPT is freely available at: https://github.com/PayamDiba/CoNSEPT.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Photon temporal-mode readout for inference of neutron star merger remnant gravitational waves

Gravitational waves emitted after neutron star binary coalescences and the information they carry about dense matter are a high-priority target for next-generation detectors. Even though such detectors are expected to observe millions of signals, detectable postmerger emission will remain rare. Here, in this work, we explore postmerger detectability and inference through an alternative detector readout scheme for data dominated by quantum-noise, which is the case above 1 kHz; photon-counting. In such a readout, signals and noise become quantized into discrete distributions corresponding to the detection of single photons measured in a chosen basis of modes. Through simulated data, we demonstrate that photon counting can be efficient even for weak signals. We find ∼1 in 100 signals with a postmerger signal-to-noise ratio of 0.2 can result in a single photon and thus be detected. Furthermore, after 2 ×10 4 signals—equivalent to 10 −2 to 1.5 years of observation—photon counting results in a twofold improvement in the measurement of the radius of a 1.6⁢𝑀 ⊙ neutron star. Constraints can be further tightened if the detector classical noise is reduced. Photon counting offers a promising alternative to traditional homodyne readout techniques for extracting information from low signal-to-noise ratio postmerger signals.

gravitational wave detection↗
DOE OSTI2024

Accurate field-level weak lensing inference for precision cosmology

We present miko, a catalog-to-cosmology pipeline for general flat-sky field-level inference, which provides access to cosmological information beyond the two-point statistics. In the context of weak lensing, we identify several new field-level analysis systematics (such as aliasing, Fourier mode-coupling, and density-induced shape noise), quantify their impact on cosmological constraints, and correct the biases to a percent level. Next, we find that model misspecification can lead to both absolute bias and incorrect uncertainty quantification for the inferred cosmological parameters in realistic simulations. The Gaussian map prior infers unbiased cosmological parameters, regardless of the true data distribution, but it yields overconfident uncertainties. The log-normal map prior quantifies the uncertainties accurately, although it requires careful calibration of the shift parameters for unbiased cosmological parameters. Here, we demonstrate systematics control down to the 2% level for both models, making them suitable for ongoing weak lensing surveys.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

I-GCN: A Graph Convolutional Network Accelerator with Runtime Locality Enhancement through Islandization

In this paper, we propose a novel hardware accelerator for GCN inference called I-GCN that significantly improves data locality and reduces unnecessary computation through a new online graph restructuring algorithm we refer to as islandization. The proposed algorithm finds clusters of nodes with strong internal but weak external connections. The islandization process yields two major benefits. First, by processing islands rather than individual nodes, there is better on-chip data reuse and fewer off-chip memory accesses. Second, there is less redundant computation as aggregation for common/shared neighbors in an island can be reused. The parallel search, identification, and leverage of graph islands are all handled purely in hardware at runtime working in an incremental pipelined manner. This is done without any preprocessing of the graph data or adjustment of the GCN model structure.

Geng, Tong↗
DOE OSTI2023

pnnl/brain_ohsu

Light sheet microscopy has made possible the 3D imaging of both fixed and live biological tissue, with samples as large as the entire mouse brain. We fine-tuned an existing model, TrailMap, using expert labeled data from axonal structures in neocortex. Without changing the network architecture, we implemented nnU-Net framework modifications in data augmentation, data foreground sampling, window learning rate, and the inference overlap method. The resulting model from these combined approaches yielded an improved F1 score

Oostrom, Marjolein↗
DOE OSTI2021

PayamDiba/CoNSEPT

Deciphering the sequence-function relationship encoded in enhancers holds the key to interpreting non-coding variants and understanding mechanisms of transcriptomic variation. Several quantitative models exist for predicting enhancer function and underlying mechanisms; however, there has been no systematic comparison of these models characterizing their relative strengths and shortcomings. Here, we interrogated a rich data set of neuroectodermal enhancers in Drosophila, representing cis- and trans- sources of expression variation, with a suite of biophysical and machine learning models. We performed rigorous comparisons of thermodynamics-based models implementing different mechanisms of activation, repression and cooperativity. Moreover, we developed a convolutional neural network (CNN) model, called CoNSEPT, that learns enhancer ‘grammar’ in an unbiased manner. CoNSEPT is the first general-purpose CNN tool for predicting enhancer function in varying conditions, such as different cell types and experimental conditions, and we show that such complex models can suggest interpretable mechanisms. We found model-based evidence for mechanisms previously established for the studied system, including cooperative activation and short-range repression. The data also favored one hypothesized activation mechanism over another and suggested an intriguing role for a direct, distance-independent repression mechanism. Our modeling shows that while fundamentally different models can yield similar fits to data, they vary in their utility for mechanistic inference.

Dibaeinia, Payam↗
DOE OSTI2024

Neural Networks to Find the Optimal Forcing for Offsetting the Anthropogenic Climate Change Effects

Abstract Of great relevance to climate engineering is the systematic relationship between the radiative forcing to the climate system and the response of the system, a relationship often represented by the linear response function (LRF) of the system. However, estimating the LRF often becomes an ill-posed inverse problem due to high-dimensionality and nonunique relationships between the forcing and response. Recent advances in machine learning make it possible to address the ill-posed inverse problem through regularization and sparse system fitting. Here, we develop a convolutional neural network (CNN) for regularized inversion. The CNN is trained using the surface temperature responses from a set of Green’s function perturbation experiments as imagery input data together with data sample densification. The resulting CNN model can infer the forcing pattern responsible for the temperature response from out-of-sample forcing scenarios. This promising proof of concept suggests a possible strategy for estimating the optimal forcing to negate certain undesirable effects of climate change. The limited success of this effort underscores the challenges of solving an inverse problem for a climate system with inherent nonlinearity. Significance Statement Predicting the climate response for a given climate forcing is a direct problem, while inferring the forcing for a given desired climate response is often an inverse, ill-posed, problem, posing a new challenge to the climate community. This study makes the first attempt to infer the radiative forcing for a given target pattern of global surface temperature response using a deep learning approach. The resulting deeply trained convolutional neural network inversion model shows promise in capturing the forcing pattern corresponding to a given surface temperature response, with a significant implication on the design of an optimal solar radiation management strategy for curbing global warming. This study also highlights the technical challenges that future research should prioritize in seeking feasible solutions to the inverse climate problem.

Ren, Huiying↗
DOE OSTI2024

Cosmology Large Angular Scale Surveyor (CLASS): 90 GHz Telescope Pointing, Beam Profile, Window Function, and Polarization Performance

The Cosmology Large Angular Scale Surveyor (CLASS) is a telescope array that observes the cosmic microwave background (CMB) over ∼75% of the sky from the Atacama Desert, Chile, at frequency bands centered near 40, 90, 150, and 220 GHz. CLASS measures the large angular scale CMB polarization to constrain the tensor-to-scalar ratio and the optical depth to last scattering. This paper presents the optical characterization of the 90 GHz telescope. Observations of the Moon establish the pointing while dedicated observations of Jupiter are used for beam calibration. The standard deviations of the pointing error in azimuth, elevation, and boresight angle are 1$^{'}_.$3, 2$^{'}_.$1, and 2$^{'}_.$0, respectively, over the first 3 yr of observations. This corresponds to a pointing uncertainty ∼7% of the beam’s full width at half-maximum (FWHM). The effective azimuthally symmetrized instrument 1D beam estimated at 90 GHz has an FWHM of 0$^°_.$620 ± 0$^°_.$003 and a solid angle of 138.7 ± 0.6(stats.) ± 1.1(sys.) μsr integrated to a radius of 4°. The corresponding beam window function drops to $b^2_ℓ$ = 0.93, 0.71, 0.14 at ℓ = 30, 100, 300, respectively. Far-sidelobes are studied using detector-centered intensity maps of the Moon and measured to be at a level of 10 −3 or below relative to the peak. The polarization angle of Tau A estimated from preliminary survey maps is 149°.6 ± 0°.2(stats.) in equatorial coordinates. The instrumental temperature-to-polarization (T → P) leakage fraction, inferred from per-detector demodulated Jupiter scan data, has a monopole component at the level of 1.7 × 10 −3 , a dipole component with an amplitude of 4.3 × 10 −3 , with no evidence of quadrupolar leakage.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

Support Vector Machines for Classification of Direct Energy Deposition Standoff Distance for Improved Process Control

A critical factor in the implementation of direct energy deposition is the ability to maintain the standoff distance between the nozzle and the build surface, as this influences powder capture efficiency and overall part quality. Due to process-related variations, layer height may vary, causing unintended variation in standoff distance and poor build quality. While prior work has utilized contact probing to qualify standoff distance during processing, in situ methods for qualification of standoff distance are of major interest. The present work seeks to understand efficacy of image-based methods for classifying standoff distance variation in real-time using support vector machines (SVMs). It was hypothesized that the size of the melt pool and the amount of spatter will have significant correlations with deviations in the standoff distance; thus, SVMs were used on a dataset that is comprised of morphological features of melt pool size and image entropy. The SVM model was used to classify melt pool images into categories according to standoff distance variation from nominal. K-folds cross validation was used to find the optimal hyperparameters for the SVM model. To understand the impact of the selected features on the classification performance and inference speed, multiple models were trained with differing numbers of included features. Results for classification score, inference time, and image preprocessing/feature extraction from these data are reported. The present results show that the SVM model was able to predict the standoff distance classification with an accuracy of 97 percent and a speed of 0.122 s per image, making it a viable solution for real-time control of standoff distance.

Klesmith, Zoe↗
DOE OSTI2023

Towards a unified nonlocal, peridynamics framework for the coarse-graining of molecular dynamics data with fractures

Molecular dynamics (MD) has served as a powerful tool for designing materials with reduced reliance on laboratory testing. However, the use of MD directly to treat the deformation and failure of materials at the mesoscale is still largely beyond reach. In this work, we propose a learning framework to extract a peridynamics model as a mesoscale continuum surrogate from MD simulated material fracture data sets. Firstly, we develop a novel coarse-graining method, to automatically handle the material fracture and its corresponding discontinuities in the MD displacement data sets. Inspired by the weighted essentially non-oscillatory (WENO) scheme, the key idea lies at an adaptive procedure to automatically choose the locally smoothest stencil, then reconstruct the coarse-grained material displacement field as the piecewise smooth solutions containing discontinuities. Then, based on the coarse-grained MD data, a two-phase optimization-based learning approach is proposed to infer the optimal peridynamics model with damage criterion. In the first phase, we identify the optimal nonlocal kernel function from the data sets without material damage to capture the material stiffness properties. Then, in the second phase, the material damage criterion is learnt as a smoothed step function from the data with fractures. As a result, a peridynamics surrogate is obtained. As a continuum model, our peridynamics surrogate model can be employed in further prediction tasks with different grid resolutions from training, and hence allows for substantial reductions in computational cost compared with MD. We illustrate the efficacy of the proposed approach with several numerical tests for the dynamic crack propagation problem in a single-layer graphene. Our tests show that the proposed data-driven model is robust and generalizable, in the sense that it is capable of modeling the initialization and growth of fractures under discretization and loading settings that are different from the ones used during training.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Density measurements for the National Ignition Facility (NIF) opacity platform

The Opacity Platform on the National Ignition Facility (NIF) has been developed to measure opacities at varying densities and temperatures relevant to the solar interior and thermal cooling rates in white dwarf stars. The typical temperatures reached at NIF range between 150 and 210 eV, which allow these measurements to be performed experimentally. The captured opacities are crucial to validating radiation-hydrodynamic models that are used in astrophysics. The NIF opacity platform has a unique new capability that allows in situ measurement of the sample expansion. The sample expansion data are used to better understand the plasma conditions in our experiments by inferring the sample density throughout the duration of the laser drive. Here, we present the details of the density measurement technique, data analysis, and recent results for Fe and MgO.

47 OTHER INSTRUMENTATION↗