Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Transfer Learning Model”

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 451 records · Page 25
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Efficient Parallelization of Irregular Applications on GPU Architectures

With the enlarging computation capacity of general Graphics Processing Units (GPUs), leveraging GPUs to accelerate parallel applications has become a critical topic in academia and industry. However, a wide range of irregular applications with the computation-/memory-intensive nature cannot easily achieve high GPU utilization. The challenges mainly involve the following aspects: first, data dependence leads to coarse-grained kernel and inefficient parallelism; second, heavy GPU memory usage may cause frequent memory evictions and extra overhead of I/O; third, specific computation patterns produce memory redundancies; last, workload balance and data reusability conjunctly benefit the overall performance, but there may exist a dynamic trade-off between them. Targeting these challenges, this dissertation proposes multiple optimizations to accelerate two real-world applications: many-body correlation functions to simulate nuclear physics in a large-scale scientific system; the other is the eALS-based matrix factorization recommendation system. To accelerate the calculations of many-body correlation functions, this dissertation presents three frameworks in GPU memory management and multi-GPU scheduling. Firstly, an optimized systematic GPU memory management framework, MemHC, utilizes a series of new memory reduction designs in GPU memory allocation, CPU/GPU communications, and GPU memory oversubscription. Secondly, an enhanced multi-GPU scheduling framework, MICCO, particularly by taking both data dimension (e.g., data reuse and data eviction) and computation dimension into account. MICCO designs a heuristic scheduling algorithm and a machine learning-based regression model to generate the optimal settings of a proposed new concept to manage the trade-off. Thirdly, a locality-aware multi-GPU scheduling framework. This scheduler leverages pipeline batch generation with a looking-ahead strategy by building local dependency graphs for memory transfer reduction and better data reuse, achieving up to 79.92% memory cost reduction and 1.67x speedup. To parallelize the eALS-based recommendation system, this dissertation proposes an efficient CPU/GPU heterogeneous recommendation system, HEALS. HEALS employs newly designed architecture-adaptive data formats to achieve load balance and good data locality on CPU and GPU. To mitigate the data dependence, HEALS presents a CPU/GPU collaboration model for both task parallelism and data parallelism with multiple kernel computation optimizations. In summary, this dissertation efficiently accelerates two typical irregular applications on GPUs by building four frameworks, including CPU/GPU collaboration, GPU memory management, and multi-GPU scheduling.

Wang, Qihan↗
DOE OSTI2021

Machine-learning-based dynamic-importance sampling for adaptive multiscale simulations

Multiscale simulations are a well-accepted way to bridge the length and time scales required for scientific studies with the solution accuracy achievable through available computational resources. Traditional approaches either solve a coarse model with selective refinement or coerce a detailed model into faster sampling, both of which have limitations. Here, we present a paradigm of adaptive, multiscale simulations that couple different scales using a dynamic-importance sampling approach. Our method uses machine learning to dynamically and exhaustively sample the phase space explored by a macro model using microscale simulations and enables an automatic feedback from the micro to the macro scale, leading to a self-healing multiscale simulation. As a result, our approach delivers macro length and time scales, but with the effective precision of the micro scale. Our approach is arbitrarily scalable as well as transferable to many different types of simulations. Overall, our method made possible a multiscale scientific campaign of unprecedented scale to understand the interactions of RAS proteins with a plasma membrane in the context of cancer research running over several days on Sierra, which is currently the second-most-powerful supercomputer in the world.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Process-informed adsorbent design guidelines for direct air capture

Direct air capture using solid adsorbents is a proven technology critical to reducing our net greenhouse gas emissions to zero and beyond. Currently, academic research into the technology mainly focuses on the development of new adsorbents. However, there is a discord between the adsorbent design and process performance. Many materials scientists focus on maximising metrics such as the CO 2 capacity of their adsorbent. Here, we combine detailed process modelling, machine learning, and extensive global sensitivity analysis, which entails varying all of the model parameters together, on a direct air capture process to show that the dry CO 2 adsorption capacity does not influence process performance for an amine-functionalised adsorbent operating in a temperature vacuum swing adsorption (TVSA) process, while it is important in a steam-assisted TVSA (S-TVSA) process. In fact, adsorption kinetics, density, and thermal conductivity are all critical attributes to obtaining a low energy penalty and reduced costs. The analysis also highlights the importance of heat transfer, directing process engineers to (alternative) adsorber designs that maximise this. By an in-depth evaluation of how process performance indicators are affected by materials properties and process operating parameters, this work provides guidance to both material scientists and process engineers towards the design of a “unicorn adsorbent” and intensified DAC processes. This will improve the performance of solid adsorbent direct air capture and help drive down the costs of this vital technology to avert the worst impacts of climate change.

42 ENGINEERING↗
DOE OSTI2025

Rocket Launch Detection with Smartphone Audio and Transfer Learning

Rocket launches generate infrasound signatures that have been detected at great distances. Due to the sparsity of the networks that have made these detections, however, most signals are detected tens of minutes to hours after the rocket launch. In this work, a method of near-real-time detection of rocket launches using data from a network of smartphones located 10–70 km from launch sites is presented. A machine learning model is trained and tested on the open-access Aggregated Smartphone Timeseries of Rocket-generated Acoustics (ASTRA), Smartphone High-explosive Audio Recordings Dataset (SHAReD), and ESC-50 datasets, resulting in a final accuracy of 97% and a false positive rate of <1%. The performance and behavior of the model are summarized, and its suitability for persistent monitoring applications is discussed.

acoustics↗
DOE OSTI2023

Uncertainty-driven dynamics for active learning of interatomic potentials

Machine learning (ML) models, if trained to data sets of high-fidelity quantum simulations, produce accurate and efficient interatomic potentials. Active learning (AL) is a powerful tool to iteratively generate diverse data sets. In this approach, the ML model provides an uncertainty estimate along with its prediction for each new atomic configuration. If the uncertainty estimate passes a certain threshold, then the configuration is included in the data set. Here we develop a strategy to more rapidly discover configurations that meaningfully augment the training data set. The approach, uncertainty-driven dynamics for active learning (UDD-AL), modifies the potential energy surface used in molecular dynamics simulations to favor regions of configuration space for which there is large model uncertainty. The performance of UDD-AL is demonstrated for two AL tasks: sampling the conformational space of glycine and sampling the promotion of proton transfer in acetylacetone. The method is shown to efficiently explore the chemically relevant configuration space, which may be inaccessible using regular dynamical sampling at target temperature conditions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Multiscale Data-Driven Seismic Full-Waveform Inversion With Field Data Study

Seismic full-waveform inversion (FWI), which uses iterative methods to estimate high-resolution subsurface models from seismograms, is a powerful imaging technique in exploration geophysics. In recent years, the computational cost of FWI has grown exponentially due to the increasing size and resolution of seismic data. Moreover, it is a nonconvex problem and can encounter local minima due to the limited accuracy of the initial velocity models or the absence of low frequencies in the measurements. To overcome these computational issues, we develop a multiscale data-driven FWI method based on fully convolutional networks (FCNs). In preparing the training data, we first develop a real-time style transform method to create a large set of synthetic subsurface velocity models from natural images. We then develop two convolutional neural networks with encoder-decoder structures to reconstruct the low- and high-frequency components of the subsurface velocity models, separately. To validate the performance of our data-driven inversion method and the effectiveness of the synthesized training set, we compare it with conventional physics-based waveform inversion approaches using both synthetic and field data. Finally, these numerical results demonstrate that, once our model is fully trained, it can significantly reduce the computation time and yield more accurate subsurface velocity models in comparison with conventional FWI.

58 GEOSCIENCES↗
DOE OSTI2021

Can artificial intelligence and data-driven machine learning models match or even replace process-driven hydrologic models for streamflow simulation?: A case study of four watersheds with different hydro-climatic regions across the CONUS

With recent developments in computational techniques, Data-driven Machine Learning Models (DMLs) have shown great potential in simulating streamflow and capturing the rainfall-runoff relationship in given watersheds, which are traditionally fulfilled by Process-based Hydrologic Models (PHMs). There are debates on whether the DMLs can outperform and possibly replace the classical PHMs for streamflow simulation and river forecasting, but no clear conclusions have been made. This study aims to investigate whether the newer DMLs have any potential in further improving the simulation accuracy of classical PHMs, and vice versa. To do this, we compared a few popular PHMs and DMLs over four watersheds across the Continental US (CONUS) that are associated with different input, climate, and regional conditions. A total of five hydrologic models were chosen, including (1) two classical lumped models, i.e., the Sacramento Soil Moisture Accounting (SAC-SMA) and Xinanjiang (XAJ); (2) one modern distributed model, termed Coupled Routing and Excess Storage (CREST); (3) and two DMLs including an Artificial Neural Networks (ANN) and a deep learning model, termed Long Short Term Memory (LSTM). Our results demonstrated that the DMLs still significantly biased when using the baseline input scenario with the PHMs. However, the DMLs fed with delayed input scenarios had great potential and can reach high simulation accuracy. The DMLs, especially the ANN, outperformed other employed models under the rainfall-runoff relationship in which rainfall dominantly drives. Furthermore, the DMLs also showed better performance in the high-flow regime, while the PHMs had a better performance for the low-flow regime, implying both PHMs and DMLs have their own merits and are worthy of joint development. In general, our study indicated a great potential of using DMLs to simulate streamflow, but further studies are still needed to verify the transferability and scalability of DMLs in large-scale experiments, such as the Distributed Model Intercomparison Projects 1&2 conducted by National Weather Services but to compare modern DMLs and PHMs.

58 GEOSCIENCES↗
DOE OSTI2022

Multimodal sensor fusion framework for residential building occupancy detection

For several years now, smart building energy systems have been a research area of intensive activity. In light of the increasing need for sustainable buildings and energy systems, this trend motivates an increasing need for a solution to reduce carbon dioxide emissions and improve energy efficiency. This work proposes a high-performing and transferable occupancy detection framework that combines sensor data from different data modalities, including time series environmental data (temperature, humidity, and illuminance), image data, and acoustic energy data using ensemble method. To draw out the best prediction performance in each modality, the proposed framework was developed, including various models that were designed to learn the occupancy patterns reflected in the physical data streams. To tackle the time series environmental data, we designed two variants of an occupancy detection spatiotemporal pattern network (Occ-STPN) that performs both feature level and decision level fusion, respectively. We also propose a new metric; the fading memory mean square error (FMMSE), that provides a fair evaluation and penalization of delayed occupancy predictions. Multiple open-sourced datasets, including the Electricity Consumption and Occupancy and the University of California, Irvine's (UCI) building occupancy detection dataset, along with our own real data collected from six different houses, were used to validate the algorithms' performance. The experimental results presented herein break down the performance for each sensing modality, and a detailed analysis of the performance is also discussed.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2026

Measurement of muon antineutrino charged current - 0 meson scattering, using the NOvA Near Detector

Antineutrino interaction cross sections are, at present, poorly constrained, particularly regarding the role of multi-nucleon processes such as 2-particle 2-hole (2p2h) interactions. The associated crosssection systematic uncertainties represent a significant challenge for precision oscillation measurements, especially for the next generation of neutrino experiments such as DUNE. We present a new measurement of the muon antineutrino charged-current cross section without mesons in the final state, using the high-statistics data set of the NOvA Near Detector. The analysis employs a cut-based selection enhanced by machine learning techniques to isolate a high-purity sample dominated by quasielastic (QE) and 2p2h interactions. We present the cross section as a function of the kinetic energy and scattering angle of the outgoing muon. We also present measurements of more model-dependent kinematic variables such as the neutrino energy and momentum transfer, to better probe the underlying nuclear physics. The results are compared against various neutrino event generators to test the robustness of current interaction models.

Vockerodt, Kevin John [Ohio State U.; Queen Mary, ↗
DOE OSTI2025

Deep Learning Advances Arctic River Water Temperature Predictions

The accelerated warming in the Arctic poses serious risks to freshwater ecosystems by altering streamflow and river thermal regimes. However, limited research on Arctic River water temperatures exists due to data scarcity and the absence of robust methodologies, which often focus on large, major river basins. To address this, we leveraged the newly released, extensive AKTEMP data set and advanced machine learning techniques to develop a Long Short-Term Memory (LSTM) model. By incorporating ERA5-Land reanalysis data and integrating physical understanding into data-driven processes, our model advanced river water temperature predictions in ungauged, snow- and permafrost-affected basins in Alaska. Our model outperformed existing approaches in high-latitude regions, achieving a median Nash-Sutcliffe Efficiency of 0.95 and root mean squared error of 1.0°C. The LSTM model learned air temperature, soil temperature, solar radiation, and thermal radiation—factors associated with energy balance—were the most important drivers of river temperature dynamics. Soil moisture and snow water equivalent were highlighted as critical factors representing key processes such as thawing, melting, and groundwater contributions. Glaciers and permafrost were also identified as important covariates, particularly in seasonal river water temperature predictions. Our LSTM model successfully captured the complex relationships between hydrometeorological factors and river water temperatures across varying timescales and hydrological conditions. This scalable and transferable approach can be potentially applied across the Arctic, offering valuable insights for future conservation and management efforts.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2026

Machine learning based prediction of airflow maldistribution in air-to-refrigerant heat exchangers

Flow maldistribution is a common challenge in heat exchanger (HX) design and particularly important for air-to-refrigerant geometries where capacity losses can approach 65%. This has a major impact on central air conditioning systems, as compact duct design motivates the use of A-type HXs which are known to be affected by airflow maldistribution. Because velocity profiles are difficult to predict, components are often oversized leading to increased material cost, system footprint, and refrigerant charge. Several studies detail airflow maldistribution for individual HXs and packages, but findings cannot always be extrapolated to new designs. In this work, a machine learning (ML) based flow profile prediction framework is developed and applied to two common package configurations: (i) A-type and (ii) U-type HXs, across a broad range of HX geometries and flow rates. Porous media CFD simulations are validated against independent data for both package types as well as comprehensive in house measurements for a finless geometry with shape optimized non-round tubes, which validates the framework for new heat transfer surfaces. The ML models are trained on the porous media CFD simulations, predicting volumetric flow rate (VFR) within 1.1% and 1.9% with maximum relative L 2 norm errors of 0.48 and 0.65, respectively, while also delivering 10 5 speed up factor compared to full porous media CFD. HX level simulations show an up to 9% reduction in heat transfer from flow maldistribution, with greater losses occurring at smaller half apex angles. This framework enables rapid and highly accurate prediction of airflow maldistribution induced capacity degradation.

42 ENGINEERING↗
DOE OSTI2026

Transfer Learning Meets Embedded Correlated Wavefunction Theory for Chemically Accurate Molecular Simulations: Application to Calcium Carbonate Ion Pairing

Achieving chemical accuracy for molecular simulations remains a central challenge in computational chemistry. Here, we present an embedded correlated wavefunction transfer learning (ECW-TL) framework for accurately simulating molecular dynamics in the condensed phase. ECW-TL incorporates high-level electron exchange and correlation effects in ECW theory while preserving the training and computational efficiency of machine-learned interatomic potentials. We demonstrate the framework on Ca 2+ –CO 3 2– ion pairing in aqueous solution, a key process underlying CO 2 mineralization in seawater. As proof of principle, we first show that fine-tuning a DFT-revPBE-D3(BJ) baseline model with embedded-DFT-SCAN data reproduces the DFT-SCAN free-energy surface within 1 kcal/mol across all solvation states. Extending the framework to embedded MP2 and localized natural-orbital CCSD(T) further refines the free-energy profile, revealing the crucial role of exact electron exchange and correlation in determining ion-pair stability and structure. The computed ion-pair association free energy is in quantitative agreement with experimental measurements, further validating the accuracy of the ECW-TL framework. ECW-TL thus provides a general, data-efficient route for transferring CW accuracy to efficient simulations of complex aqueous and interfacial chemical processes.

cluster chemistry↗
DOE OSTI2022

PlasmidHostFinder: Prediction of Plasmid Hosts Using Random Forest

Plasmids play a major role facilitating the spread of antimicrobial resistance between bacteria. Understanding the host range and dissemination trajectories of plasmids is critical for surveillance and prevention of antimicrobial resistance. Identification of plasmid host ranges could be improved using automated pattern detection methods compared to homology-based methods due to the diversity and genetic plasticity of plasmids. In this study, we developed a method for predicting the host range of plasmids using machine learning—specifically, random forests. We trained the models with 8,519 plasmids from 359 different bacterial species per taxonomic level; the models achieved Matthews correlation coefficients of 0.662 and 0.867 at the species and order levels, respectively. Our results suggest that despite the diverse nature and genetic plasticity of plasmids, our random forest model can accurately distinguish between plasmid hosts. This tool is available online through the Center for Genomic Epidemiology (https://cge.cbs.dtu.dk/services/PlasmidHostFinder/).

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Instance Segmentation for Direct Measurements of Satellites in Metal Powders and Automated Microstructural Characterization from Image Data

In this work, we propose instance segmentation as a useful tool for image analysis in materials science. Instance segmentation is an advanced technique in computer vision which generates individual segmentation masks for every object of interest that is recognized in an image. Using an out-of-the-box implementation of Mask R-CNN, instance segmentation is applied to images of metal powder particles produced through gas atomization. Leveraging transfer learning allows for the analysis to be conducted with a very small training set of labeled images. As well as providing another method for measuring the particle size distribution, we demonstrate the first direct measurements of the satellite content in powder samples. After analyzing the results for the labeled data dataset, the trained model was used to generate measurements for a much larger set of unlabeled images. The resulting particle size measurements showed reasonable agreement with laser scattering measurements. The satellite measurements were self-consistent and showed good agreement with the expected trends for different samples. Finally, we present a small case study showing how instance segmentation can be used to measure spheroidite content in the UltraHigh Carbon Steel DataBase, demonstrating the flexibility of the technique.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Machine learning nonequilibrium electron forces for spin dynamics of itinerant magnets

Abstract We present a generalized potential theory for conservative as well as nonconservative forces for the Landau-Lifshitz magnetization dynamics. Importantly, this formulation makes possible an elegant generalization of the Behler-Parrinello machine learning (ML) approach, which is a cornerstone of ML-based quantum molecular dynamics methods, to the modeling of force fields in adiabatic spin dynamics of out-of-equilibrium itinerant magnetic systems. We demonstrate our approach by developing a deep-learning neural network that successfully learns the electron-mediated exchange fields in a driven s-d model computed from the nonequilibrium Green’s function method. We show that dynamical simulations with forces predicted from the neural network accurately reproduce the voltage-driven domain-wall propagation. Our work also lays the foundation for ML modeling of spin transfer torques and opens a avenue for ML-based multi-scale modeling of nonequilibrium dynamical phenomena in itinerant magnets and spintronics.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Alpert multi-wavelets for functional inverse problems: direct optimization and deep learning

Computational engineering models often contain unknown entities (e.g. parameters, initial and boundary conditions) that require estimation from other measured observable data. Estimating such unknown entities is challenging when they involve spatio-temporal fields because such functional variables often require an infinite-dimensional representation. Here, we address this problem by transforming an unknown functional field using Alpert wavelet bases and truncating the resulting spectrum. Hence the problem reduces to the estimation of few coefficients that can be performed using common optimization methods. We apply this method on a one-dimensional heat transfer problem where we estimate the heat source field varying in both time and space. The observable data is comprised of temperature measured at several thermocouples in the domain. This latter is composed of either copper or stainless steel. The optimization using our method based on wavelets is able to estimate the heat source with an error between 5% and 7%. We analyze the effect of the domain material and number of thermocouples as well as the sensitivity to the initial guess of the heat source. Finally, we estimate the unknown heat source using a different approach based on deep learning techniques where we consider the input and output of a multi-layer perceptron in wavelet form. We find that this deep learning approach is more accurate than the optimization approach with errors below 4%.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Machine learning to accelerate screening for Marcus reorganization energies

Understanding and predicting the charge transport properties of π-conjugated materials is an important challenge for designing new organic electronic devices, such as solar cells, plastic transistors, light-emitting devices, and chemical sensors. A key component of the hopping mechanism of charge transfer in these materials is the Marcus reorganization energy which serves as an activation barrier to hole or electron transfer. While modern density functional methods have proven to accurately predict trends in intramolecular reorganization energy, such calculations are computationally expensive. In this work, we outline active machine learning methods to predict computed intramolecular reorganization energies of a wide range of polythiophenes and their use toward screening new compounds with low internal reorganization energies. Our models have an overall root mean square error (RMSE) of ±0.113 eV, but a much smaller RMSE of only ±0.036 eV on the new screening set. Since the larger error derives from high-reorganization energy compounds, the new method is highly effective to screen for compounds with potentially efficient charge transport parameters.

14 SOLAR ENERGY↗
DOE OSTI2025

Modeling Equilibrium Solid–Liquid Interfaces under Effective Constant Chemical Potential Using Machine Learning Interatomic Potentials

The chemical potential (μ) of species in solution is essential for understanding various chemical processes at interfaces. Molecular dynamics (MD) simulations, constrained by fixed compositions, cannot maintain constant chemical potential with reference to a targeted concentration or chemical potential under nonequilibrium or dynamic conditions, as solute species can migrate to the interface and deplete (or enrich) the bulk due to solute-interface interactions. In this study, we introduce a simple and computationally efficient approach named iterative quasi-constant chemical potential molecular dynamics (iqCμMD) simulation, which helps simulate targeted molar concentrations of species in solution. iqCμMD overcomes the limitations of conventional MD by adjusting the number of species in the solution to reach a target bulk concentration (chemical potential), which allows simulation of the interface under the bulk conditions comparable to experiment. We demonstrate our approach using machine learning interatomic potential (MLIP)-based MD simulations of the Na 2 SO 4,aq –graphene interface, and to show the transferability of our approach, we also perform classical force field-based MD simulations of NaCl aq –air and NaCl aq –graphite interfaces, which produce comparable results to previous CμMD simulations. Our results also show that the iqCμMD approach efficiently achieves the desired bulk ion concentration within two iterations, and by utilizing MLIPs, we can achieve converged results using relatively small-scale simulations compared to previous CμMD simulations. By combining iqCμMD with MLIP-driven simulations, solid–liquid interfaces can be modeled under an effective constant chemical potential with DFT-level accuracy. Here, we show that iqCμMD offers a robust and simple computational framework for constant chemical potential simulations, as its only requirement is to be able to converge interfacial simulations with a measurable bulk region.

Chemical structure↗