Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Model Predictive Control”

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 1,279 records · Page 71
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Fuel molecular structure effect on autoignition of highly branched iso-alkanes at low-to-intermediate temperatures: Iso-octane versus iso-dodecane

Highly branched iso-alkanes are an important class of hydrocarbons found in conventional petroleum-derived and alternative renewable fuels used for combustion applications. Recognizing that chemical kinetics for most of these iso-alkanes, especially at low-to-intermediate temperatures, has not been well studied, an experimental and modeling investigation of two selected iso-alkanes, iso-octane (2,2,4-trimethylpentane, iC8) and iso-dodecane (2,2,4,6,6-pentamethylheptane, iC12), is conducted to understand the fuel molecular structure effect on their autoignition characteristics. Using a rapid compression machine (RCM), the ignition responses of iC8 and iC12 at varying pressures, temperatures, and equivalence ratios are characterized and compared. The newly-acquired experimental ignition delay times have been compared with the literature RCM and shock tube data, demonstrating the complementary nature of the current dataset. Further comparison of the experimental pressure traces and ignition delay times illustrates the reactivity crossover between iC8 and iC12. Namely, there exists a temperature window in the negative temperature coefficient regime within which iC12 is less reactive than iC8, but iC12 becomes more reactive outside this temperature window. Furthermore, a chemical kinetic model of iso-alkanes including both iC8 and iC12 is developed. Simulated results using this model are then compared to the experimental data obtained in this study and available in the literature, showing its ability to predict the experimental trends. Chemical kinetic analyses have also been conducted to identify the important reaction pathways controlling autoignition at varying conditions, and to elucidate the underlying mechanism leading to different reactivity trends between iC8 and iC12.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Two decades of fumigation data from the Soybean Free Air Concentration Enrichment facility

Abstract The Soybean Free Air Concentration Enrichment (SoyFACE) facility is the longest running open-air carbon dioxide and ozone enrichment facility in the world. For over two decades, soybean, maize, and other crops have been exposed to the elevated carbon dioxide and ozone concentrations anticipated for late this century. The facility, located in East Central Illinois, USA, exposes crops to different atmospheric concentrations in replicated octagonal ~280 m 2 Free Air Concentration Enrichment (FACE) treatment plots. Each FACE plot is paired with an untreated control (ambient) plot. The experiment provides important ground truth data for predicting future crop productivity. Fumigation data from SoyFACE were collected every four seconds throughout each growing season for over two decades. Here, we organize, quality control, and collate 20 years of data to facilitate trend analysis and crop modeling efforts. This paper provides the rationale for and a description of the SoyFACE experiments, along with a summary of the fumigation data and collation process, weather and ambient data collection procedures, and explanations of air pollution metrics and calculations.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Drift kinetic effects on plasma response to resonant magnetic perturbation for EU DEMO design

Abstract A systematic investigation of the plasma response to an externally applied resonant magnetic perturbation (RMP) field, for the purpose of controlling edge localized modes, is carried out for an EU DEMO reference plasma. Particular emphasis is placed on the role of kinetic effects associated with both thermal particles and fusion-born alphas. The single fluid, resistive model predicts a large peak amplification of the n = 1 ( n is the toroidal mode number) plasma response to the target equilibrium, which is found to be close to the Troyon no-wall limit. A more advanced response model, including kinetic resonances between the RMP perturbation and drift motions of thermal and energetic particles, on the other hand, finds a strong suppression of the n = 1 field amplification. A major role is played by the precessional drift resonance of fusion-born alphas. A strong parallel sound wave damping model is found to well reproduce the full kinetic response results for the DEMO plasma, in terms of both the resonant field response amplitude and the plasma displacement. Finally, both fluid and kinetic models produce similar responses for the n = 2 and 3 RMP fields for the considered DEMO plasma, whilst kinetic effects again become important for the n = 4 RMP due to proximity of the reference plasma to the no-wall limit for the n = 4 ideal kink instability.

Physics↗
DOE OSTI2025

Optimization of simulated high-field side lower hybrid current drive coupling using machine learning predictions of scrape-off layer density

Lower hybrid current drive (LHCD) is a potential source of non-inductive off-axis current drive (CD) for tokamaks. Although LHCD has been successfully deployed on a number of tokamaks, it is highly sensitive to the scrape-off layer (SOL) conditions local to the LHCD launcher. Large gaps between the launcher and plasma core, SOL turbulence, or edge density perturbations due to edge-localized modes can hamper CD or cause large reflected power. These coupling issues in part motivated the installation of an LHCD launcher on the high-field side (HFS) of DIII-D. On the HFS, the SOL is less turbulent and more controllable compared to the low-field side. This quiescence may result in more predictable edge conditions and thus a more predictable CD. Here, in this work, HFS SOL reflectometry measurements are predicted from global plasma parameters using machine learning models. The SOL predictions coupled with the full-wave simulation of the LHCD launcher allow for the prediction of reflected power, directivity, and arcing risk before the discharge. Launcher performance is then optimized using multi-objective Bayesian optimization, finding the shot parameters that result in an optimal SOL density that maximizes CD while minimizing the risk of arcing. The predictions and optimizations of LHCD performance are then accelerated using a surrogate model of the full-wave LHCD simulation.

Bayesian optimization↗
DOE OSTI2024

In-Situ Species Concentration Measurements In Ammonia-Mix Flames Using Ftir Spectroscopy

Hydrogen and ammonia represent two carbon-free fuel sources that could be used in place of current fossil energy sources in combustion systems. To develop optimized ammonia combustion systems, validated modeling tools are needed. In the open literature, it has been shown that the complex chemistry associated with fuel-bound nitrogen contained in ammonia differs greatly from natural gas or hydrogen combustion. As a result, several new chemical kinetic mechanisms have been developed. Many of these mechanisms have been validated experimentally, however this has primarily focused on bulk parameters such as laminar flame speed and ignition delay time. Critically, high quality measurements of species concentrations are needed under controlled conditions which are easily represented by simple models. In this paper, direct, in-situ measurements of species concentrations and gas temperature are performed in a laminar flat-flame burner. This arrangement enables comparison with 1D model predictions, better isolating chemical kinetics from the fluid dynamics. Quantitative species concentrations are determined by absorption spectroscopy using an FTIR spectrometer. Fuel compositions representative of cracked ammonia (NH 3 /H 2 ) and ammonia-natural gas (NH 3 /CH 4 ) are considered for rich and lean equivalence ratios. A major focus of the paper is on the selection of spectral features for nitric oxide and ammonia and correcting for large amounts of baseline H 2 O absorption.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Relationships between Immersion Freezing and Crystal Habit for Arctic Mixed-Phase Clouds—A Numerical Study

Abstract The number concentration of ice particles in Arctic mixed-phase clouds is a major controlling factor of cloud lifetime. The relationships between ice nucleation mode and ice crystal habit development are not yet constrained by observations. This study uses a habit-predicting microphysical scheme within a 3D large-eddy simulation model to evaluate the relationship between immersion freezing and ice habit in a simulated Arctic mixed-phase cloud case. Three immersion freezing parameterizations are considered: a volume-dependent freezing scheme (VF), a parameterization limited to activated droplets (C-AC), and a parameterization limited to coarse aerosol particles (C-CM). Both C-AC and C-CM are based on classical nucleation theory. The freezing rate with VF is found to be greater in downdraft regions than in updraft regions due to the downdraft having a higher number concentration of large droplets. The C-AC cases show active freezing of small droplets near cloud top, whereas in the C-CM cases, mainly the 8–32- μ m-sized droplets freeze in updraft regions near the cloud base. Because the initial crystal size is assumed to affect the axis ratio of hexagonal plates, the VF cases produce crystals with larger axis ratios, resulting in smaller mode radii than the C-AC cases. In all cases, irregular polycrystals dominate near cloud top and a band-like structure develops within the cloud, which qualitatively agrees with previous observations. In the VF and C-CM cases, unactivated large droplets arising from coarse-mode aerosol particles contributed significantly to the freezing rate, producing an important influence on crystal habit.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Layer Time Control for Large Scale Additive Manufacturing Using High Performance Computing

This work proposes to optimize an additive manufacturing AM process to reduce energy and printing cost. The polymer AM process is inherently dependent on the time-temperature history of each layer to maintain geometric tolerances and mechanical integrity. Our preliminary study shows that regression-based layer time control model using thermal images could result in up to 30% build time reduction for simple geometries. This proposed work would use high-performance computing (HPC) to couple the data-driven model with thermal simulation for better predicting layer temperature profiles, improving throughput of large-scale additive manufacturing, and reducing its energy cost. We have developed a method to optimize a layer deposition time (a.k.a. layer time) for large-scale AM via physics-based simulations. A long layer time leads to an over-cooled surface on which a new layer is deposited, and therefore, it may result in a weak bonding or debonding between layers, cracking, or warping. A short layer time leads to a high temperature of the structure due to insufficient cooling, and therefore, the structure may not be stiff enough and may collapse during manufacturing. Therefore, it is important to estimate the optimal layer time in additive manufacturing for a high-quality product. The temperature of a top layer right before deposition is recommended to be slightly higher than the glass temperature of the material. A temperature cooling was approximated to an exponential function of time, and the optimized layer time was obtained based on a target temperature while maintaining a minimal printing time. The material used is carbon fiber-reinforced polycarbonate (CF/PC), and the large-scale deposition system used is LSAM TM from Thermwood Corporation. Three different layer time cases were used for experiments, and a series of thermal images were obtained via an infra-red (IR) camera during the entire AM processes. AM process simulations were performed using a finite element method and the temperature profiles from the simulation were in good agreements with those from experiments. The layer time optimization was performed based on the temperature profiles from the simulations. A layer temperature with the optimal layer time was confirmed as the target temperature through simulation. In addition to the development of a layer time optimization method, we have developed a numerical framework for AM simulation with element activations in sync with toolpath, based on an open source finite element framework, DEAL.II. A major portion of this work was presented at SAMPE 2022 Conference and Exhibition on May 2022, and published in Proceedings of SAMPE 2022.

42 ENGINEERING↗
DOE OSTI2023

NGEE Arctic Rainfall Simulator Validation Data from Los Alamos National Laboratory, New Mexico, Summer 2022

Experiments evaluating the uniformity and intensity of rainfall produced by the NGEE Arctic Rainfall Simulator (NARS) were conducted at Los Alamos National Laboratory, New Mexico, over summer 2022. Petri dishes were placed in a grid within the NARS plot. Simulated rainfall was collected in each petri dish and the intensity and uniformity of the simulator was subsequently calculated. This data package contains two .csv files, one that summarizes the rainfall intensity and uniformity for each experiment, the other that contains individual petri dish water volume and intensity for each plot location and experiment. The Python scripts to control NARS are also included. The NGEE Arctic Rainfall Simulator (NARS) is a variable intensity rainfall simulator (RFS) with a frame design based on the Humphry et al. (2002) RFS and a water delivery system based on the Walnut Gulch (Paige et al., 2004) RFS. The NARS uses an aluminum frame that is fully deconstructable for transportation to field locations and a water system that enables variable rain intensity. Rain intensity control and data collection are automated using a Raspberry Pi microcomputer. The Next-Generation Ecosystem Experiments: Arctic (NGEE Arctic), was a 15-year research effort (2012-2027) to reduce uncertainty in Earth System Models by developing a predictive understanding of carbon-rich Arctic ecosystems and feedbacks to climate. NGEE Arctic was supported by the Department of Energy's Office of Biological and Environmental Research. The NGEE Arctic project had two field research sites: 1) located within the Arctic polygonal tundra coastal region on the Barrow Environmental Observatory (BEO) and the North Slope near Utqiagvik (Barrow), Alaska and 2) multiple areas on the discontinuous permafrost region of the Seward Peninsula north of Nome, Alaska. Through observations, experiments, and synthesis with existing datasets, NGEE Arctic provided an enhanced knowledge base for multi-scale modeling and contributed to improved process representation at global pan-Arctic scales within the Department of Energy's Earth system Model (the Energy Exascale Earth System Model, or E3SM), and specifically within the E3SM Land Model component (ELM).

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

In-Situ Species Concentration Measurements in Ammonia-Mix Flames Using FTIR Spectroscopy

Hydrogen and ammonia represent two carbon-free fuel sources that could be used in place of current fossil energy sources in combustion systems. To develop optimized ammonia combustion systems, validated modeling tools are needed. In the open literature, it has been shown that the complex chemistry associated with fuel-bound nitrogen contained in ammonia differs greatly from natural gas or hydrogen combustion. As a result, several new chemical kinetic mechanisms have been developed. Many of these mechanisms have been validated experimentally, however this has primarily focused on bulk parameters such as laminar flame speed and ignition delay time. Critically, high quality measurements of species concentrations are needed under controlled conditions which are easily represented by simple models. In this paper, direct, in-situ measurements of species concentrations and gas temperature are performed in a laminar flat-flame burner. This arrangement enables comparison with 1D model predictions, better isolating chemical kinetics from the fluid dynamics. Quantitative species concentrations are determined by absorption spectroscopy using an FTIR spectrometer. Fuel compositions representative of cracked ammonia (NH3/H2) and ammonia-natural gas (NH3/CH4) are considered for rich and lean equivalence ratios. A major focus of the paper is on the selection of spectral features for nitric oxide and ammonia and correcting for large amounts of baseline H2O absorption.

Bedick, Clinton↗
DOE OSTI2022

Neural network accelerator for quantum control

Efficient quantum control is necessary for practical quantum computing implementations with current technologies. Conventional algorithms for determining optimal control parameters are computationally expensive, largely excluding them from use outside of the simulation. Existing hardware solutions structured as lookup tables are imprecise and costly. By designing a machine learning model to approximate the results of traditional tools, a more efficient method can be produced. Such a model can then be synthesized into a hardware accelerator for use in quantum systems. In this study, we demonstrate a machine learning algorithm for predicting optimal pulse parameters. This algorithm is lightweight enough to fit on a low-resource FPGA and perform inference with a latency of 175 ns and pipeline interval of 5 ns with > 0.99 gate fidelity. In the long term, such an accelerator could be used near quantum computing hardware where traditional computers cannot operate, enabling quantum control at a reasonable cost at low latencies without incurring large data bandwidths outside of the cryogenic environment.

43 PARTICLE ACCELERATORS↗
DOE OSTI2021

Predicting Nugget Size of Resistance Spot Welds Using Infrared Thermal Videos With Image Segmentation and Convolutional Neural Network

Resistance spot welding (RSW) is a widely adopted joining technique in automotive industry. Recent advancement in sensing technology makes it possible to collect thermal videos of the weld nugget during RSW using an infrared (IR) camera. The effective and timely analysis of such thermal videos has the potential of enabling in situ nondestructive evaluation (NDE) of the weld nugget by predicting nugget thickness and diameter. Deep learning (DL) has demonstrated to be effective in analyzing imaging data in many applications. However, the thermal videos in RSW present unique data-level challenges that compromise the effectiveness of most pre-trained DL models. We propose a novel image segmentation method for handling the RSW thermal videos to improve the prediction performance of DL models in RSW. The proposed method transforms raw thermal videos into spatial-temporal instances in four steps: video-wise normalization, removal of uninformative images, watershed segmentation, and spatial-temporal instance construction. The extracted spatial-temporal instances serve as the input data for training a DL-based NDE model. The proposed method is able to extract high-quality data with spatial-temporal correlations in the thermal videos, while being robust to the impact of unknown surface emissivity. Overall, our case studies demonstrate that the proposed method achieves better prediction of nugget thickness and diameter than predicting without the transformation.

42 ENGINEERING↗
DOE OSTI2022

Investigation of Multiple Data Streams for Gearbox Bearing Fault Prediction Through Machine-Learning Models

Operations and maintenance (O&M) cost of wind plant accounts up to 30% of total energy cost, which can be reduced through continuous monitoring and successfully detecting incipient wind turbine failures. To accomplish this, condition monitoring and predictive maintenance systems are being implemented in wind industry to support O&M decision making. A wide range of approaches for condition monitoring and fault prediction have been developed. These approaches generally use historical data of wind turbines collected by Supervisory Control and Data Acquisition (SCADA) system to identify patterns that lead to failure. These SCADA data show the overall condition of a wind turbine and can be leveraged to detect when the turbine's performance is degrading and to identify if a fault is developing. However, it becomes challenging to predict the failure of a specific wind turbine gearbox bearing, because the SCADA data are often not directly linked to the component. To bridge the gap, we have investigated features calculated from SCADA data using physics-based models and the gearbox design over the years. The damaged metric we used in the physics domain is frictional energy. Combining these physics domain variables with SCADA data as inputs to various machine learning models for gearbox bearing fault prediction, we have demonstrated the benefits of leveraging both physics and data domain models. It was an attempt to improve frictional-energy-based damage metric by adding data domain inputs, as we had learned that the frictional-energy-based damage metric alone is not sufficient to single out failed bearings from healthy. As condition monitoring data (either vibration or oil debris data) has become available at more and more wind plants, we would like to evaluate whether by adding the condition monitoring data can help further improve the performance of frictional-energy-based damage metric for gearbox bearing fault prediction. Both cases by modeling through various machine learning algorithms are discussed in this study along with some observations.

fault prediction↗
DOE OSTI2022

Symmetry-breaking in patch formation on triangular gold nanoparticles by asymmetric polymer grafting

Synthesizing patchy particles with predictive control over patch size, shape, placement and number has been highly sought-after for nanoparticle assembly research, but is fraught with challenges. Here we show that polymers can be designed to selectively adsorb onto nanoparticle surfaces already partially coated by other chains to drive the formation of patchy nanoparticles with broken symmetry. In our model system of triangular gold nanoparticles and polystyrene-b-polyacrylic acid patch, single- and double-patch nanoparticles are produced at high yield. These asymmetric single-patch nanoparticles are shown to assemble into self-limited patch-patch connected bowties exhibiting intriguing plasmonic properties. To unveil the mechanism of symmetry-breaking patch formation, we develop a theory that accurately predicts our experimental observations at all scales—from patch patterning on nanoparticles, to the size/shape of the patches, to the particle assemblies driven by patch-patch interactions. Both the experimental strategy and theoretical prediction extend to nanoparticles of other shapes such as octahedra and bipyramids. Our work provides an approach to leverage polymer interactions with nanoscale curved surfaces for asymmetric grafting in nanomaterials engineering.

42 ENGINEERING↗
DOE OSTI2024

Monovalent ion–graphene oxide interactions are controlled by carboxylic acid groups: Sum frequency generation spectroscopy studies

Graphene oxide (GO) is a two-dimensional, mechanically strong, and chemically tunable material for separations. Elucidating GO–ion–water interactions at the molecular scale is highly important for predictive understanding of separation systems. However, direct observations of the nanometer region by GO surfaces under operando conditions are not trivial. Therefore, thin films of GO at the air/water interface can be used as model systems. With this approach, we study the effects of alkali metal ions on water organization near graphene oxide films at the air/water interface using vibrational sum frequency generation (SFG) spectroscopy. Here, we also use an arachidic acid Langmuir monolayer as a benchmark for a pure carboxylic acid surface. Theoretical modeling of the concentration-dependent sum frequency signal from graphene oxide and arachidic acid surfaces reveals that the adsorption of monovalent ions is mainly controlled by the carboxylic acid groups on graphene oxide. An in-depth analysis of sum frequency spectra reveals at least three distinct water populations with different hydrogen bonding strengths. The origin of each population can be identified from concentration dependent variations of their SFG signal. Interestingly, an interfacial water structure seemed mostly insensitive to the character of the alkali cation, in contrast to similar studies conducted at the silica/water interface. However, we observed an ion-specific effect with lithium, whose strong hydration prevented direct interactions with the graphene oxide film.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Progress in the validation of rotor aerodynamic codes using field data

Within the framework of the fourth phase of the International Energy Agency (IEA) Wind Task 29, a large comparison exercise between measurements and aeroelastic simulations has been carried out featuring three simulation cases in axial, sheared and yawed inflow conditions. Results were obtained from more than 19 simulation tools originating from 12 institutes, ranging in fidelity from blade element momentum (BEM) to computational fluid dynamics (CFDs) and compared to state-of-the-art field measurements from the 2 MW DanAero turbine. More than 15 different variable types ranging from lifting-line variables to blade surface pressures, loads and velocities have been compared for the different conditions, resulting in over 250 comparison plots. The result is a unique insight into the current status and accuracy of rotor aerodynamic modeling. For axial flow conditions, a good agreement was found between the various code types, where a dedicated grid sensitivity study was necessary for the CFD simulations. However, compared to wind tunnel experiments on rotors featuring controlled conditions, it remains a challenge to achieve good agreement of absolute levels between simulations and measurements in the field. For sheared inflow conditions, uncertainties due to rotational and unsteady effects on airfoil data result in the CFD predictions standing out above the codes that need input of sectional airfoil data. However, it was demonstrated that using CFD-synthesized airfoil data is an effective means to bypass this shortcoming. For yawed flow conditions, it was observed that modeling of the skewed wake effect is still problematic for BEM codes where CFD and free vortex wake codes inherently model the underlying physics correctly. The next step is a comparison in turbulent inflow conditions, which is featured in IEA Wind Task 47. Doing this analysis in cooperation under the auspices of the IEA Wind Technology Collaboration Program (TCP) has led to many mutual benefits for the participants. The large size of the consortium brought ample manpower for the analysis where the learning process by combining several complementary experiences and modeling techniques gave valuable insights that could not be found when the analysis is carried out individually.

17 WIND ENERGY↗
DOE OSTI2022

Improbability of Nuclear Criticality in Compacted Criticality Control Overpacks after Room Closure by Salt Creep at Waste Isolation Pilot Plant

Based on the rationale presented, nuclear criticality is improbable after salt creep causes compaction of criticality control overpacks (CCOs) disposed at the Waste Isolation Pilot Plant, an operating repository in bedded salt for the disposal of transuranic (TRU) waste from atomic energy defense activities. For most TRU waste, the possibility of post-closure criticality is exceedingly small either because the salt neutronically isolates TRU waste canisters or because closure of a disposal room from salt creep does not sufficiently compact the low mass of fissile material. The criticality potential has been updated here because of the introduction of CCOs, which may dispose up to 380 fissile gram equivalent plutonium-239 in each container. The criticality potential is evaluated through high-fidelity geomechanical modeling of a disposal room filled with CCOs during two representative conditions: (1) large salt block fall, and (2) gradual salt compaction (without brine seepage and subsequent gas generation to permit maximum room closure). Geomechanical models of rock fall demonstrate three tiers of CCOs are not greatly disrupted. Geomechanical models of gradual room closure from salt creep predict irregular arrays of closely packed CCOs after 1000 years, when room closure has asymptotically approached maximum compaction. Criticality models of spheres and cylinders of 380 fissile gram equivalent of plutonium (as oxide) at the predicted irregular spacing demonstrate that an array of CCOs is not critical when surrounded by salt and magnesium oxide, provided the amount of hydrogenous material shipped in the CCO (usually water and plastics) is controlled or boron carbide (a neutron poison) is mixed with the fissile contents.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2020

Prediction of histone post-translational modifications using deep learning

Abstract Motivation Histone post-translational modifications (PTMs) are involved in a variety of essential regulatory processes in the cell, including transcription control. Recent studies have shown that histone PTMs can be accurately predicted from the knowledge of transcription factor binding or DNase hypersensitivity data. Similarly, it has been shown that one can predict PTMs from the underlying DNA primary sequence. Results In this study, we introduce a deep learning architecture called DeepPTM for predicting histone PTMs from transcription factor binding data and the primary DNA sequence. Extensive experimental results show that our deep learning model outperforms the prediction accuracy of the model proposed in Benveniste et al. (PNAS 2014) and DeepHistone (BMC Genomics 2019). The competitive advantage of our framework lies in the synergistic use of deep learning combined with an effective pre-processing step. Our classification framework has also enabled the discovery that the knowledge of a small subset of transcription factors (which are histone-PTM and cell-type-specific) can provide almost the same prediction accuracy that can be obtained using all the transcription factors data. Availabilityand implementation https://github.com/dDipankar/DeepPTM. Supplementary information Supplementary data are available at Bioinformatics online.

Baisya, Dipankar Ranjan (ORCID:0000000267847359)↗
DOE OSTI2022

GrainNN: A neighbor-aware long short-term memory network for predicting microstructure evolution during polycrystalline grain formation

High fidelity simulations of grain formation in alloys are an indispensable tool for process-to-mechanical-properties characterization. Such simulations, however, can be computationally expensive as they require fine spatial and temporal discretizations. Their cost becomes an obstacle to parametric studies and ensemble runs and ultimately makes downstream tasks like optimal control and uncertainty quantification challenging. To enable such downstream tasks, we introduce GrainNN, an efficient and accurate reduced-order model for epitaxial grain growth in additive manufacturing conditions. GrainNN is a sequence-to-sequence long-short-term-memory (LSTM) deep neural network that evolves the dynamics of manually crafted features. Its innovations are (1) an attention mechanism with grain-microstructure-specific transformer architecture; and (2) an overlapping combination of several clones of the network to generalize to grain configurations that are different from those used for training. This design enables GrainNN to predict grain formation for unseen physical parameters, grain number, domain size and geometry. Furthermore, GrainNN not only reconstructs the quantities of interest but also can be pointwise accurate. In our numerical experiments, we use a polycrystalline phase field method to both generate the training data and assess GrainNN. For multiparametric, ensemble simulations with many grains, GrainNN can be orders of magnitude faster than phase field simulations, while delivering 5%–15% pointwise error. Additionally, this speedup includes the cost of the phase field simulations for generating training data.

36 MATERIALS SCIENCE↗