Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Differentiability”

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 883 records · Page 49
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

Differential Wavevector Distribution of Surface-Enhanced Raman Scattering and Fluorescence in a Film-Coupled Plasmonic Nanowire Cavity

Here, we report on the experimental observation of differential wavevector distribution of surface-enhanced Raman scattering (SERS) and fluorescence from dye molecules confined to a gap between plasmonic silver nanowire and a thin, gold mirror. The fluorescence was mainly confined to higher values of in-plane wavevectors, whereas SERS signal was uniformly distributed along all the wavevectors. The optical energy momentum spectra from the distal end of the nanowire revealed strong polarization dependence of this differentiation. All these observations were corroborated by full-wave three-dimensional numerical simulations, which further revealed an interesting connection between out-coupled wavevectors and parameters such as hybridized modes in the gap-plasmon cavity, and orientation and location of molecular dipoles in the geometry. Our results reveal a new prospect of discriminating electronic and vibrational transitions in resonant dye molecules using a subwavelength gap plasmonic cavity in the continuous-wave excitation limit, and can be further harnessed to engineer molecular radiative relaxation processes in momentum space.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

Strong Differential Monovalent Anion Selectivity in Narrow Diameter Carbon Nanotube Porins

Inner pores of carbon nanotubes combine extremely fast water transport and ion selectivity that could potentially be useful for high-performance water desalination and separation applications. In this study, we used dye-quenching halide assays and stopped-flow spectrometry to determine intrinsic permeability of three small monovalent halide anions (chloride, bromide, iodide) and one pseudohalide anion (thiocyanate) through narrow 0.8 nm-diameter carbon nanotube porins (CNTPs). These measurements revealed unexpectedly strong differential ion selectivity with permeabilities of different ions varying by up to two orders of magnitude. Removal of the negative charge from the nanotube entrance increased anion permeability by only a relatively small factor, indicating that electrostatic repulsion was not a major determinant of CNTP selectivity. First principle molecular dynamics simulations revealed that the origin of this strong differential ion selectivity is partial dehydration of anions upon entry into the narrow CNTP channels.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Differentiable, Learnable, Regionalized Process-Based Models With Multiphysical Outputs can Approach State-Of-The-Art Hydrologic Prediction Accuracy

Predictions of hydrologic variables across the entire water cycle have significant value for water resources management as well as downstream applications such as ecosystem and water quality modeling. Recently, purely data-driven deep learning models like long short-term memory (LSTM) showed seemingly insurmountable performance in modeling rainfall runoff and other geoscientific variables, yet they cannot predict untrained physical variables and remain challenging to interpret. Here, we show that differentiable, learnable, process-based models (called δ models here) can approach the performance level of LSTM for the intensively observed variable (streamflow) with regionalized parameterization. We use a simple hydrologic model HBV as the backbone and use embedded neural networks, which can only be trained in a differentiable programming framework, to parameterize, enhance, or replace the process-based model's modules. Without using an ensemble or post-processor, δ models can obtain a median Nash-Sutcliffe efficiency of 0.732 for 671 basins across the USA for the Daymet forcing data set, compared to 0.748 from a state-of-the-art LSTM model with the same setup. For another forcing data set, the difference is even smaller: 0.715 versus 0.722. Meanwhile, the resulting learnable process-based models can output a full set of untrained variables, for example, soil and groundwater storage, snowpack, evapotranspiration, and baseflow, and can later be constrained by their observations. Both simulated evapotranspiration and fraction of discharge from baseflow agreed decently with alternative estimates. The general framework can work with models with various process complexity and opens up the path for learning physics from big data.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Inferring Plant Acclimation and Improving Model Generalizability With Differentiable Physics‐Informed Machine Learning of Photosynthesis

Net photosynthesis (A N ) is a key component of the global carbon cycle influencing climate feedback over decadal scales. Although plant acclimation to environmental changes can modify A N , traditional vegetation models in Earth system models (ESMs) often rely on plant functional type (PFT)-specific parameterizations or simplified acclimation assumptions limiting generalizability across time, space, and PFTs. In this study, we developed a differentiable photosynthesis model to learn the environmental dependencies of V c,max25 (maximum carboxylation rate at 25°C, representing photosynthetic capacity), as this genre of hybrid physics-informed machine learning can seamlessly train neural networks and process-based equations together. Compared to PFT-specific parameterization of V c,max25 , learning the environment dependencies of key photosynthetic parameters improved model spatiotemporal generalizability. Applying environmental acclimation to V c,max25 led to substantial variations in global mean A N indicating the need to address acclimation in ESMs. The model effectively captured multivariate observations (V c,max25 , A N , and stomatal conductance (g s )) simultaneously with multivariate constraints, improving generalization across space and PFTs. It also learned sensible acclimation relationships of V c,max25 to different environmental conditions. The model explained more than 54%, 57%, and 62% of the variance of A N , g s , and V c,max25 , respectively, presenting a first global-scale spatial test benchmark of A N and g s . These results highlight the potential for differentiable modeling to enhance process-based modules in ESMs and effectively leverage information from large, multivariate data sets.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Sulfate differentially stimulates but is not respired by diverse anaerobic methanotrophic archaea

Abstract Sulfate-coupled anaerobic oxidation of methane (AOM) is a major methane sink in marine sediments. Multiple lineages of anaerobic methanotrophic archaea (ANME) often coexist in sediments and catalyze this process syntrophically with sulfate-reducing bacteria (SRB), but the potential differences in ANME ecophysiology and mechanisms of syntrophy remain unresolved. A humic acid analog, anthraquinone 2,6-disulfonate (AQDS), could decouple archaeal methanotrophy from bacterial sulfate reduction and serve as the terminal electron acceptor for AOM (AQDS-coupled AOM). Here in sediment microcosm experiments, we examined variations in physiological response between two co-occurring ANME-2 families (ANME-2a and ANME-2c) and tested the hypothesis of sulfate respiration by ANME-2. Sulfate concentrations as low as 100 µM increased AQDS-coupled AOM nearly 2-fold matching the rates of sulfate-coupled AOM. However, the SRB partners remained inactive in microcosms with sulfate and AQDS and neither ANME-2 families respired sulfate, as shown by their cellular sulfur contents and anabolic activities measured using nanoscale secondary ion mass spectrometry. ANME-2a anabolic activity was significantly higher than ANME-2c, suggesting that ANME-2a was primarily responsible for the observed sulfate stimulation of AQDS-coupled AOM. Comparative transcriptomics showed significant upregulation of ANME-2a transcripts linked to multiple ABC transporters and downregulation of central carbon metabolism during AQDS-coupled AOM compared to sulfate-coupled AOM. Surprisingly, genes involved in sulfur anabolism were not differentially expressed during AQDS-coupled AOM with and without sulfate amendment. Collectively, this data indicates that ANME-2 archaea are incapable of respiring sulfate, but sulfate availability differentially stimulates the growth and AOM activity of different ANME lineages.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2018

Optimization of carbon and energy utilization through differential translational efficiency

Control of translation is vital to all species. Here we employ a multi-omics approach to decipher condition-dependent translational regulation in the model acetogen Clostridium ljungdahlii . Integration of data from cells grown autotrophically or heterotrophically revealed that pathways critical to carbon and energy metabolism are under strong translational regulation. Major pathways involved in carbon and energy metabolism are not only differentially transcribed and translated, but their translational efficiencies are differentially elevated in response to resource availability under different growth conditions. We show that translational efficiency is not static and that it changes dynamically in response to mRNA expression levels. mRNAs harboring optimized 5'-untranslated region and coding region features, have higher translational efficiencies and are significantly enriched in genes encoding carbon and energy metabolism. In contrast, mRNAs enriched in housekeeping functions harbor sub-optimal features and have lower translational efficiencies. As a result, we propose that regulation of translational efficiency is crucial for effectively controlling resource allocation in energy-deprived microorganisms.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2019

Perturbed myoepithelial cell differentiation in BRCA mutation carriers and in ductal carcinoma in situ

Myoepithelial cells play key roles in normal mammary gland development and in limiting pre-invasive to invasive breast tumor progression, yet their differentiation and perturbation in ductal carcinoma in situ (DCIS) are poorly understood. Here, we investigated myoepithelial cells in normal breast tissues of BRCA1 and BRCA2 germline mutation carriers and in non-carrier controls, and in sporadic DCIS. We found that in the normal breast of non-carriers, myoepithelial cells frequently co-express the p63 and TCF7 transcription factors and that p63 and TCF7 show overlapping chromatin peaks associated with differentiated myoepithelium-specific genes. In contrast, in normal breast tissues of BRCA1 mutation carriers the frequency of p63 + TCF7 + myoepithelial cells is significantly decreased and p63 and TCF7 chromatin peaks do not overlap. These myoepithelial perturbations in normal breast tissues of BRCA1 germline mutation carriers may play a role in their higher risk of breast cancer. The fraction of p63 + TCF7 + myoepithelial cells is also significantly decreased in DCIS, which may be associated with invasive progression.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Learning emergent partial differential equations in a learned emergent space

We propose an approach to learn effective evolution equations for large systems of interacting agents. This is demonstrated on two examples, a well-studied system of coupled normal form oscillators and a biologically motivated example of coupled Hodgkin-Huxley-like neurons. For such types of systems there is no obvious space coordinate in which to learn effective evolution laws in the form of partial differential equations. In our approach, we accomplish this by learning embedding coordinates from the time series data of the system using manifold learning as a first step. In these emergent coordinates, we then show how one can learn effective partial differential equations, using neural networks, that do not only reproduce the dynamics of the oscillator ensemble, but also capture the collective bifurcations when system parameters vary. The proposed approach thus integrates the automatic, data-driven extraction of emergent space coordinates parametrizing the agent dynamics, with machine-learning assisted identification of an emergent PDE description of the dynamics in this parametrization.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Ecological trade-offs drive phenotypic and genetic differentiation of Arabidopsis thaliana in Europe

Plant diversity is shaped by trade-offs between traits related to competitive ability, propagule dispersal, and stress resistance. However, we still lack a clear understanding of how these trade-offs influence species distribution and population dynamics. In Arabidopsis thaliana , recent genetic analyses revealed a group of cosmopolitan genotypes that successfully recolonized Europe from its center after the last glaciation, excluding older (relict) lineages from the distribution except for their north and south margins. Here, we tested the hypothesis that cosmopolitans expanded due to higher colonization ability, while relicts persisted at the margins due to higher tolerance to competition and/or stress. We compared the phenotypic and genetic differentiation between 71 European genotypes originating from the center, and the south and north margins. We showed that a trade-off between plant fecundity and seed mass shapes the differentiation of A. thaliana in Europe, suggesting that the success of the cosmopolitan groups could be explained by their high dispersal ability. However, at both north and south margins, we found evidence of selection for alleles conferring low dispersal but highly competitive and stress-resistance abilities. This study sheds light on the role of ecological trade-offs as evolutionary drivers of the distribution and dynamics of plant populations.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2017

Differential electron emission from polycyclic aromatic hydrocarbon molecules under fast ion impact

Interaction between polycyclic aromatic hydrocarbon (PAH) molecule and energetic ion is a subject of interest in different areas of modern physics. Here, we present measurements of energy and angular distributions of absolute double differential electron emission cross section for coronene (C 24 H 12 ) and fluorene (C 13 H 10 ) molecules under fast bare oxygen ion impact. For coronene, the angular distributions of the low energy electrons are quite different from that of simpler targets like Ne or CH 4 , which is not the case for fluorene. The behaviour of the higher electron energy distributions for both the targets are similar to that for simple targets. In case of coronene, a clear signature of plasmon resonance is observed in the analysis of forward-backward angular asymmetry of low energy electron emission. For fluorene, such signature is not identified probably due to lower oscillator strength of plasmon compared to the coronene. The theoretical calculation based on the first-order Born approximation with correct boundary conditions (CB1), in general, reproduced the experimental observations qualitatively, for both the molecules, except in the low energy region for coronene, which again indicates the role of collective excitation. Single differential and total cross sections are also deduced. An overall comparative study is presented.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2017

Negative Differential Conductance & Hot-Carrier Avalanching in Monolayer WS2 FETs

The high field phenomena of inter-valley transfer and avalanching breakdown have long been exploited in devices based on conventional semiconductors. In this Article, we demonstrate the manifestation of these effects in atomically-thin WS 2 field-effect transistors. The negative differential conductance exhibits all of the features familiar from discussions of this phenomenon in bulk semiconductors, including hysteresis in the transistor characteristics and increased noise that is indicative of travelling high-field domains. It is also found to be sensitive to thermal annealing, a result that we attribute to the influence of strain on the energy separation of the different valleys involved in hot-electron transfer. This idea is supported by the results of ensemble Monte Carlo simulations, which highlight the sensitivity of the negative differential conductance to the equilibrium populations of the different valleys. At high drain currents (>10 μA/μm) avalanching breakdown is also observed, and is attributed to trap-assisted inverse Auger scattering. This mechanism is not normally relevant in conventional semiconductors, but is possible in WS 2 due to the narrow width of its energy bands. The various results presented here suggest that WS 2 exhibits strong potential for use in hot-electron devices, including compact high-frequency sources and photonic detectors.

36 MATERIALS SCIENCE↗
DOE OSTI2017

The Role of Chromatin Density in Cell Population Heterogeneity during Stem Cell Differentiation

We incorporate three-dimensional (3D) conformation of chromosome (Hi-C) and single-cell RNA sequencing data together with discrete stochastic simulation, to explore the role of chromatin reorganization in determining gene expression heterogeneity during development. While previous research has emphasized the importance of chromatin architecture on activation and suppression of certain regulatory genes and gene networks, our study demonstrates how chromatin remodeling can dictate gene expression distribution by folding into distinct topological domains. We hypothesize that the local DNA density during differentiation accentuate transcriptional bursting due to the crowding effect of chromatin. This phenomenon yields a heterogeneous cell population, thereby increasing the potential of differentiation of the stem cells.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2019

Analog simulator of integro-differential equations with classical memristors

An analog computer makes use of continuously changeable quantities of a system, such as its electrical, mechanical, or hydraulic properties, to solve a given problem. While these devices are usually computationally more powerful than their digital counterparts, they suffer from analog noise which does not allow for error control. We will focus on analog computers based on active electrical networks comprised of resistors, capacitors, and operational amplifiers which are capable of simulating any linear ordinary differential equation. However, the class of nonlinear dynamics they can solve is limited. In this work, by adding memristors to the electrical network, we show that the analog computer can simulate a large variety of linear and nonlinear integro-differential equations by carefully choosing the conductance and the dynamics of the memristor state variable. We study the performance of these analog computers by simulating integro-diferential models related to fluid dynamics, nonlinear Volterra equations for population growth, and quantum models describing non-Markovian memory effects, among others. Finally, we perform stability tests by considering imperfect analog components, obtaining robust solutions with up to 13% relative error for relevant timescales.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

Systematic Comparison of High-throughput Single-Cell and Single-Nucleus Transcriptomes during Cardiomyocyte Differentiation

A comprehensive reference map of all cell types in the human body is necessary for improving our understanding of fundamental biological processes and in diagnosing and treating disease. High-throughput single-cell RNA sequencing techniques have emerged as powerful tools to identify and characterize cell types in complex and heterogeneous tissues. However, extracting intact cells from tissues and organs is often technically challenging or impossible, for example in heart or brain tissue. Single-nucleus RNA sequencing provides an alternative way to obtain transcriptome profiles of such tissues. To systematically assess the differences between high-throughput single-cell and single-nuclei RNA-seq approaches, we compared Drop-seq and DroNc-seq, two microfluidic-based 3' RNA capture technologies that profile total cellular and nuclear RNA, respectively, during a time course experiment of human induced pluripotent stem cells (iPSCs) differentiating into cardiomyocytes. Clustering of timeseries transcriptomes from Drop-seq and DroNc-seq revealed six distinct cell types, five of which were found in both techniques. Furthermore, single-cell trajectories reconstructed from both techniques reproduced expected differentiation dynamics. We then applied DroNc-seq to postmortem heart tissue to test its performance on heterogeneous human tissue samples. Our data confirm that DroNcseq yields similar results to Drop-seq on matched samples and can be successfully used to generate reference maps for the human cell atlas.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

The inheritance of anthracnose (Colletotrichum sublineola) resistance in sorghum differential lines QL3 and IS18760

Anthracnose caused by the fungal pathogen C. sublineola is an economically important constraint on worldwide sorghum production. The most effective strategy to safeguard yield is through the introgression of resistance alleles. This requires elucidation of the genetic basis of the different resistance sources that have been identified. In this study, 223 recombinant inbred lines (RILs) derived from crossing anthracnose-differentials QL3 (96 RILs) and IS18760 (127 RILs) with the common susceptible parent PI609251 were evaluated at four field locations in the United States (Florida, Georgia, Texas, and Puerto Rico) for their anthracnose resistance response. Both RIL populations were highly susceptible to anthracnose in Florida and Georgia, while in Puerto Rico and Texas they were segregating for anthracnose resistance response. A genome scan using a composite linkage map of 982 single nucleotide polymorphisms (SNPs) detected two genomic regions of 4.31 and 0.85 Mb on chromosomes 4 and 8, respectively, that explained 10–27% of the phenotypic variation in Texas and Puerto Rico. In parallel, a subset of 43 RILs that contained 67% of the recombination events were evaluated against anthracnose pathotypes from Arkansas (2), Puerto Rico (2) and Texas (4) in the greenhouse. A genome scan showed that the 7.57 Mb region at the distal end of the short arm of chromosome 5 is associated with the resistance response against the pathotype AMP-048 from Arkansas. Comparative analysis identified the genomic region on chromosome 4 overlaps with an anthracnose resistance locus identified in another anthracnose-differential line, SC414-12E, indicating this genomic region is of interest for introgression in susceptible sorghum germplasm. Candidate gene analysis for the resistance locus on chromosome 5 identified an R -gene cluster that has high similarity to another R -gene cluster associated with anthracnose resistance on chromosome 9.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Physics-informed machine learning with differentiable programming for heterogeneous underground reservoir pressure management

Abstract Avoiding over-pressurization in subsurface reservoirs is critical for applications like CO $$_2$$ 2 sequestration and wastewater injection. Managing the pressures by controlling injection/extraction are challenging because of complex heterogeneity in the subsurface. The heterogeneity typically requires high-fidelity physics-based models to make predictions on CO $$_2$$ 2 fate. Furthermore, characterizing the heterogeneity accurately is fraught with parametric uncertainty. Accounting for both, heterogeneity and uncertainty, makes this a computationally-intensive problem challenging for current reservoir simulators. To tackle this, we use differentiable programming with a full-physics model and machine learning to determine the fluid extraction rates that prevent over-pressurization at critical reservoir locations. We use DPFEHM framework, which has trustworthy physics based on the standard two-point flux finite volume discretization and is also automatically differentiable like machine learning models. Our physics-informed machine learning framework uses convolutional neural networks to learn an appropriate extraction rate based on the permeability field. We also perform a hyperparameter search to improve the model’s accuracy. Training and testing scenarios are executed to evaluate the feasibility of using physics-informed machine learning to manage reservoir pressures. We constructed and tested a sufficiently accurate simulator that is 400 000 times faster than the underlying physics-based simulator, allowing for near real-time analysis and robust uncertainty quantification.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Physics-embedded inverse analysis with algorithmic differentiation for the earth’s subsurface

Abstract Inverse analysis has been utilized to understand unknown underground geological properties by matching the observational data with simulators. To overcome the underconstrained nature of inverse problems and achieve good performance, an approach is presented with embedded physics and a technique known as algorithmic differentiation. We use a physics-embedded generative model, which takes statistically simple parameters as input and outputs subsurface properties (e.g., permeability or P-wave velocity), that embeds physical knowledge of the subsurface properties into inverse analysis and improves its performance. We tested the application of this approach on four geologic problems: two heterogeneous hydraulic conductivity fields, a hydraulic fracture network, and a seismic inversion for P-wave velocity. This physics-embedded inverse analysis approach consistently characterizes these geological problems accurately. Furthermore, the excellent performance in matching the observational data demonstrates the reliability of the proposed method. Moreover, the application of algorithmic differentiation makes this an easy and fast approach to inverse analysis when dealing with complicated geological structures.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Deep transfer operator learning for partial differential equations under conditional shift

Transfer learning enables the transfer of knowledge gained while learning to perform one task (source) to a related but different task (target), hence addressing the expense of data acquisition and labelling, potential computational power limitations and dataset distribution mismatches. Here, we propose a new transfer learning framework for task-specific learning (functional regression in partial differential equations) under conditional shift based on the deep operator network (DeepONet). Task-specific operator learning is accomplished by fine-tuning task-specific layers of the target DeepONet using a hybrid loss function that allows for the matching of individual target samples while also preserving the global properties of the conditional distribution of the target data. Inspired by conditional embedding operator theory, we minimize the statistical distance between labelled target data and the surrogate prediction on unlabelled target data by embedding conditional distributions onto a reproducing kernel Hilbert space. We demonstrate the advantages of our approach for various transfer learning scenarios involving nonlinear partial differential equations under diverse conditions due to shifts in the geometric domain and model dynamics. Our transfer learning framework enables fast and efficient learning of heterogeneous tasks despite considerable differences between the source and target domains.

42 ENGINEERING↗