SEARCH · Engineering Papers

Results for “deep generative models”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 757 records · Page 42

All sources NASA NTRS DOE OSTI

DOE OSTI2026

Knowledge-guided graph machine learning for spatially distributed prediction of daily discharge and nitrogen export dynamics

Spatially distributed prediction of streamflow and nitrogen export dynamics is essential for precision management of agricultural watersheds. While temporal deep learning models such as Long Short-Term Memory (LSTM) have shown strong performance at basin scales, their ability to generalize spatially is limited by insufficient representation of spatial dependencies and flow paths, particularly under data-scarce conditions. To address this gap, we propose HydroGraphNet, a knowledge-guided graph machine learning framework that integrates process-based knowledge and explicit spatial learning into temporal modeling. This framework incorporates directed graph topology to encode watershed connectivity and upstream inflows, with mass balance constraints to improve physical consistency. To enhance generalization in sparsely monitored regions, HydroGraphNet is pretrained on synthetic data generated by the SWAT+ (Soil and Water Assessment Tool Plus) model. We evaluated HydroGraphNet in the Upper Sangamon River Basin (44 HUC-12 subwatersheds, 2001–2020) against two LSTM baselines: a lumped basin-level model and a distributed variant. When benchmarked on SWAT+ simulations in pretraining, HydroGraphNet improved test NSEs by 8.9% (discharge) and 13.7% (NO₃–N load) in temporal extrapolation, and by 27.1% and 34.7% in spatial extrapolation, relative to the Lumped LSTM baseline. After fine-tuning with USGS monitoring data, the model achieved mean test NSE (KGE) scores of 0.768 (0.861) for discharge and 0.626 (0.664) for NO₃–N load, substantially outperforming baselines. Attribution analysis further highlighted the importance of upstream inflow representation and graph-based spatial learning in capturing cross-subwatershed dependencies. The model also reproduced seasonal hydrological and biogeochemical patterns consistent with known processes, demonstrating its robustness and process fidelity for spatially distributed prediction. Altogether, HydroGraphNet advances the integration of physical knowledge and spatially explicit learning in hydrological modeling, offering a generalizable framework for distributed modeling to support spatially targeted water quality management in data-scarce watersheds.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2024

Quantum fluctuations lead to glassy electron dynamics in the good metal regime of electron doped KTaO 3

One of the central challenges in condensed matter physics is to comprehend systems that have strong disorder and strong interactions. In the strongly localized regime, their subtle competition leads to glassy electron dynamics which ceases to exist well before the insulator-to-metal transition is approached as a function of doping. Here, we report on the discovery of glassy electron dynamics deep inside the good metal regime of an electron-doped quantum paraelectric system: KTaO 3 . We reveal that upon excitation of electrons from defect states to the conduction band, the excess injected carriers in the conduction band relax in a stretched exponential manner with a large relaxation time, and the system evinces simple aging phenomena—a telltale sign of glassy dynamics. Most significantly, we observe a critical slowing down of carrier dynamics below 35 K, concomitant with the onset of quantum paraelectricity in the undoped KTaO 3 . Our combined investigation using second harmonic generation technique, density functional theory and phenomenological modeling demonstrates quantum fluctuation-stabilized soft polar modes as the impetus for the glassy behavior. This study addresses one of the most fundamental questions regarding the potential promotion of glassiness by quantum fluctuations and opens a route for exploring glassy dynamics of electrons in a well-delocalized regime.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2024

Event generator for jet tomography in electron-ion collisions

We develop the first event generator, the electron-heavy-ion-jet-interaction-generator (eHIJING), for the jet tomography study of electron-ion collisions. In this generator, energetic jet partons produced from the initial hard scattering undergo multiple collisions with the nuclear target. The collision rate is proportional to the transverse-momentum-dependent (TMD) gluon density in the nucleus, which is given by a simple model inspired by the physics of gluon saturation. Medium-modified QCD splitting functions within the higher-twist (HT) and generalized higher-twist (GHT) frameworks are utilized to simulate parton showering in the nuclear medium that takes into account the non-Abelian Landau-Pomeranchuck-Midgal interference effect. Employing eHIJING, we revisit hadron production in semi-inclusive deep inelastic scattering (SIDIS) as measured by EMC, HERMES, and recent CLAS experiments. eHIJING with both GT and GHT frameworks gives reasonably good descriptions of these experimental data. Predictions for experiments at the future electron-ion colliders are also provided. It is demonstrated that future measurements of the transverse momentum broadening of single hadron spectra can be used to map out the two-dimensional kinematic (Q 2 , x B ) dependence of the jet transport coefficient $\hat{q}$ in cold nuclear matter.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2023

Predicting future well performance for environmental remediation design using deep learning

Here in this study, we developed a deep learning (DL) framework with a multi-channel three-dimensional convolutional neural network (MC3D-CNN) to predict well performance and thereby assist future environmental remediation design. Such prediction of extraction well performance at designated locations is critical for configuring pump-and-treat (P&T) well network design and operation, setting reasonable target closure dates for overall remedying, and estimating remedy costs. The framework is developed with operational and monitoring data routinely collected during P&T remedy operations, including well extraction and injection rates as well as in situ contaminant concentrations. Traditionally, the collected data were rarely used for purposes other than assessing past well performance and the accuracy of the conceptual site model. However, recent advances in data-driven computational approaches enable better use of the large datasets to inform future well performance, enhance site characterization, and improve remediation planning. In this study, we established a DL framework to integrate transient three-dimensional contaminant plumes and multiple aquifer properties (e.g., hydraulic conductivity and hydrostratigraphic maps) to identify characteristic patterns controlling and representing extraction well mass recovery, aiming at providing future mass recovery estimates for existing wells and candidate wells at any proposed locations. We evaluated our framework by using a realistic synthetic dataset generated from a well-calibrated flow and transport model used in the 200 West Area of the U.S. Department of Energy’s Hanford Site in southeastern Washington state. The multi-channel feature in our framework allows integration of various types and temporal densities of training datasets for DL model development. Overall, we found that the trained DL model achieved an accuracy of over 90% in ranking extraction well performance in validation datasets, and over 80% in predicting high-performance-ranking well locations. This data-informed approach provides a flexible tool to support adaptive site management, streamline decision-making, and potentially reduce remediation time and costs. Our DL framework can be used as a filtering tool to improve the current P&T network optimization design by reducing the number of candidate well locations.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

Search for a new $Z$' gauge boson in 4$μ$ events with the ATLAS experiment

This paper presents a search for a new Z' vector gauge boson with the ATLAS experiment at the Large Hadron Collider using pp collision data collected at $\sqrt{s}$ = 13 TeV, corresponding to an integrated luminosity of 139 fb -1 . The new gauge boson Z' is predicted by Lμ - Lτ models to address observed phenomena that cannot be explained by the Standard Model. The search examines the four-muon (4μ) final state, using a deep learning neural network classifier to separate the Z' signal from the Standard Model background events. The di-muon invariant masses in the 4μ events are used to extract the Z' resonance signature. No significant excess of events is observed over the predicted background. Upper limits at a 95% confidence level on the Z' production cross-section times the decay branching fraction of pp → Z'μμ → 4μ are set from 0.31 to 4.3 fb for the Z' mass ranging from 5 to 81 GeV. The corresponding common coupling strengths, g Z' , of the Z' boson to the second and third generation leptons above 0.003 – 0.2 have been excluded.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

DOE OSTI2025

Digital Real-Time Simulation and Power Quality Analysis of a Hydrogen-Generating Nuclear-Renewable Integrated Energy System

This paper investigates the challenges and solutions associated with integrating a hydrogen-generating nuclear-renewable integrated energy system (NR-IES) under a transactive energy framework. The proposed system directs excess nuclear power to hydrogen production during periods of low grid demand while utilizing renewables to maintain grid stability. Using digital real-time simulation (DRTS) in the Typhoon HIL 404 model, the dynamic interactions between nuclear power plants, electrolyzers, and power grids are analyzed to mitigate issues such as harmonic distortion, power quality degradation, and low power factor caused by large non-linear loads. A three-phase power conversion system is modeled using the Typhoon HIL 404 model and includes a generator, a variable load, an electrolyzer, and power filters. Active harmonic filters (AHFs) and hybrid active power filters (HAPFs) are implemented to address harmonic mitigation and reactive power compensation. The results reveal that the HAPF topology effectively balances cost efficiency and performance and significantly reduces active filter current requirements compared to AHF-only systems. During maximum electrolyzer operation at 4 MW, the grid frequency dropped below 59.3 Hz without filtering; however, the implementation of power filters successfully restored the frequency to 59.9 Hz, demonstrating its effectiveness in maintaining grid stability. Future work will focus on integrating a deep reinforcement learning (DRL) framework with real-time simulation and optimizing real-time power dispatch, thus enabling a scalable, efficient NR-IES for sustainable energy markets.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2020

Video Summarization Using Deep Action Recognition Features and Robust Principal Components Analysis

In an instance where desired pre-defined actions, behaviors, or other categories are known a priori, various video classification and recognition models can be trained to discover those classifications and their location within the video. Absent that information, one might still be tasked with identifying interesting portions within a video, a process which—if done manually—is onerous and time-consuming as it requires manual inspection of the video itself. Recognizing high-level interesting segments within a whole video has been a general area of interest due to the ubiquity of video data. However the size of the data makes storage, retrieval, and inspection of large collections of videos cumbersome. This problem motivates the task of generating shortened clips highlighting the primary content of a video, relieving the burden of having to watch the entire video. This paper presents an unsupervised method of creating shortened clips of videos, enabling the rapid review of the most interesting content within a video. Our method uses features extracted from pre-trained action recognition models as input to online moving window robust principal component analysis to generate summaries. The procedure is tested on a publicly available video summarization dataset and demonstrates comparable performance to state-of-the-art in an un-augmented setting while requiring no training.

Claborne, Daniel M.↗

DOE OSTI2020

Well-to-Wheels Analysis of Zero-Emission Plug-In Battery Electric Vehicle Technology for Medium- and Heavy-Duty Trucks

Conventional diesel medium- and heavy-duty vehicles (MHDVs) create large amount of air emissions. With the advancement in technology and reduction in the cost of batteries, plug-in battery electric vehicles (BEVs) are increasingly attractive options for improving energy efficiency and reducing air emissions of MHDVs. In this paper, we compared the well-to-wheels (WTW) greenhouse gases (GHGs) and criteria air pollutant emissions of MHD BEVs with their conventional diesel counterparts across weight classes and vocations. We expanded the Greenhouse gases, Regulated Emissions, and Energy use in Technologies (GREET) model to conduct the WTW analysis of MHDVs. The fuel economy for a wide range of MHDV weight classes and vocations, over various driving cycles, was evaluated using a high-fidelity vehicle dynamic simulation software (Autonomie). The environmental impacts of MHD BEVs are sensitive to the source of electricity used to recharge their batteries. The WTW results show that MHD BEVs significantly improve environmental sustainability of MHDVs by providing deep reductions in WTW GHGs, nitrogen oxides, volatile organic compounds, and carbon monoxide emissions, compared to conventional diesel counterparts. Increasing shares of renewable and natural gas technologies in future national and regional electricity generation are expected to reduce WTW particulate matters and sulfur oxide emissions for further improvement of the environmental performance of MHD BEVs.

33 ADVANCED PROPULSION SYSTEMS↗

DOE OSTI2022

Deep Learning Based Frequency Stability Assessment in Power Grid with High Renewables

Frequency stability assessment is one critical aspect of power system security assessment. Traditional N-1 screening method is based on the simulations of a few typical daily and seasonal operation scenarios. However, the increasing integration of inverter-based renewables and the retirement of conventional synchronous generators result in decreasing system inertia and growing complexity of system operating conditions. Selecting a few typical operation scenarios cannot cover all operating conditions, and the time-domain simulation of all operation conditions requires tremendous time. This paper proposes a more efficient frequency stability assessment method based on deep learning. The affinity propagation clustering algorithm is used to divide the dataset into different clusters, so the selected dataset for training can cover the diversified operating conditions as much as possible. Also, feature normalization is applied to both the training dataset and testing dataset in order to remove any unnecessary bias. Especially, trained model based on full dataset normalization has bounded error in the prediction. The case study on the reduced 240-bus WECC system demonstrates that the proposed method can predict accurate frequency nadir with limited training dataset. The deep learning model using the revised feature normalization can predict more accurate frequency nadir than that using the traditional feature normalization and has very small maximum prediction error.

affinity propagation↗

DOE OSTI2024

A deep decarbonization framework for the United States economy – a sector, sub-sector, and end-use based approach

Achieving the United States' target of net-zero greenhouse gas emissions by 2050 will require technological transformations and energy sector mitigation. To understand the role of dynamically evolving technologies, identify synergies and dissonance and the effect of allocating limited low-carbon biomass resources in decarbonizing the U.S. economy, we developed the Decarbonization Scenario Analysis Model. A Life Cycle Assessment based approach is implemented considering the U.S. economy as the functional unit, to estimate greenhouse gas mitigation potential for projected energy demand based on several sector-level and cross-sectoral decarbonization pathways. Direct and supply-chain emissions are accounted, resulting from changes in patterns of energy generation and consumption, technology breakthroughs, and reductions in fugitive emissions over time at the granularity of economic sectors, sub-sectors, and end-use. Decarbonization strategies are implemented over a reference case developed using Energy Information Administration (EIA AEO) projection of economic activities for 2020–2050. Based on the considered scenarios, 80–90% economy-wide decarbonization relative to the 2020 reference case is projected. Electrification, low-carbon fuels, and reduction of fugitive emissions play the most significant role to decarbonization. The majority of the remaining emissions are accounted to the supply-chain and end-use emissions from natural gas and diesel fossil-based fuels in heavy duty transportation and heavy industries, highlighting the need for developing low-carbon and carbon-negative alternatives to mitigate those fossil-based carbon emissions.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

DOE OSTI2023

Predicting the synthesizability of crystalline inorganic materials from the data of known material compositions

Abstract Reliably identifying synthesizable inorganic crystalline materials is an unsolved challenge required for realizing autonomous materials discovery. In this work, we develop a deep learning synthesizability model ( SynthNN ) that leverages the entire space of synthesized inorganic chemical compositions. By reformulating material discovery as a synthesizability classification task, SynthNN identifies synthesizable materials with 7× higher precision than with DFT-calculated formation energies. In a head-to-head material discovery comparison against 20 expert material scientists, SynthNN outperforms all experts, achieves 1.5× higher precision and completes the task five orders of magnitude faster than the best human expert. Remarkably, without any prior chemical knowledge, our experiments indicate that SynthNN learns the chemical principles of charge-balancing, chemical family relationships and ionicity, and utilizes these principles to generate synthesizability predictions. The development of SynthNN will allow for synthesizability constraints to be seamlessly integrated into computational material screening workflows to increase their reliability for identifying synthetically accessible materials.

36 MATERIALS SCIENCE↗

DOE OSTI2022

Developing a Decision Support System for Regional Agricultural Nonpoint Salinity Pollution Management: Application to the San Joaquin River, California

Environmental problems and production losses associated with irrigated agriculture, such as salinity, degradation of receiving waters, such as rivers, and deep percolation of saline water to aquifers, highlight water-quality concerns that require a paradigm shift in resource-management policy. New tools are needed to assist environmental managers in developing sustainable solutions to these problems, given the nonpoint source nature of salt loads to surface water and groundwater from irrigated agriculture. Equity issues arise in distributing responsibility and costs to the generators of this source of pollution. This paper describes an alternative approach to salt regulation and control using the concept of “Real-Time Water Quality management”. The approach relies on a continually updateable WARMF (Watershed Analysis Risk Management Framework) forecasting model to provide daily estimates of salt load assimilative capacity in the San Joaquin River and assessments of compliance with salinity concentration objectives at key monitoring sites on the river. The results of the study showed that the policy combination of well-crafted river salinity objectives by the regulator and the application of an easy-to use and maintain decision support tool by stakeholders have succeeded in minimizing water quality (salinity) exceedances over a 20-year study period.

real-time management economics↗

DOE OSTI2022

Automatic Generation of High-Performance Convolution Kernels on ARM CPUs for Deep Learning

In this work, we present FastConv, a template-based code auto-generation open source library that can automatically generate high-performance deep learning convolution kernels of arbitrary matrices/tensors shapes. FastConv is based on the Winograd algorithm, which is reportedly the highest performing algorithm for the time-consuming convolution layers of convolutional neural networks. ARM CPUs cover a wide range designs and specifications, from embedded devices to HPC-grade CPUs. The leads to the dilemma of how to consistently optimize Winograd-based convolution solvers for convolution layers of different shapes. FastConv addresses this problem by using templates to auto-generate multiple shapes of tuned kernels variants suitable for skinny tall matrices. As a performance portable library, FastConv transparently searches for the best combination of kernel shapes, cache tiles, scheduling of loop orders, packing strategies, access patterns, and online/offline computations. Auto-tuning is used to search the parameter configuration space for the best performance for a given target architecture and problem size. The experiments with layer-wise evaluation on the VGG--16 model confirms a 1.25x performance gains is got by tuning the Winograd library. Integrated comparison results shows 1.02x to 1.40x, 1.14x to 2.17x, and 1.22x and 2.48x speedup is achieved over NNPACK, Arm NN, and FeatherCNN on the Kunpeng 920 beside few cases. Furthermore, problem size performance portability experiments with various convolution shapes shows that FastConv achieves 1.2x to 1.7x speedup and 2x to 22x speedup over NNPACK and ARM NN inference engine using Winograd on Kunpeng 920 . CPU performance portability evaluation on the VGG--16 show an average speedup over NNPACK of 1.42x, 1.21x, 1.26x, 1.37x, 2.26x, and 11.02x is observed on Kunpeng 920, Snapdragon 835, 855, 888, Apple M1, and AWS Graviton2, respectively.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Physics informed neural network can retrieve rate and state friction parameters from acoustic monitoring of laboratory stick-slip experiments

Various machine learning (ML) and deep learning (DL) techniques have been recently applied to the forecasting of laboratory earthquakes from friction experiments. The magnitude and timing of shear failures in stick-slip cycles are predicted using features extracted from the recorded ultrasonic or acoustic emission (AE) signals. In addition, the Rate and State Friction (RSF) constitutive laws are extensively used to model the frictional behavior of faults. In this work, we use data from shear experiments coupled with passive acoustic (variance, kurtosis, and AE rate) interleaved with active source ultrasonic monitoring (transmitted wave amplitude) to develop physics-informed neural network (PINN) models incorporating the RSF law and AE rate generation equation with wave amplitude serving as a proxy for friction state variable. This PINN framework allows learning RSF parameters from stick-slip experiments rather than measuring them through a series of velocity step experiments. We observe that when the stick-slip cycles are irregular, the PINN models outperform the data-driven DL models. Transfer learning (TL) PINN models are also developed by pre-training on data collected at one normal stress level followed by forecasting shear failures and retrieving RSF parameters at other stress levels (i.e., with different recurrence intervals) after retraining on a limited amount of new data. Our findings suggest that TL models perform better compared to standalone models. Both standalone and TL PINN-estimated RSF parameters and their ground truth values show excellent agreements thus demonstrating that RSF parameters can be retrieved from laboratory stick-slip experiments using the corresponding acoustic data and that the transmitted wave amplitude provides a good representation of the evolving frictional state during stick-slips.

58 GEOSCIENCES↗

DOE OSTI2023

Pantropical Indo-Atlantic temperature gradient modulates multi-decadal AMOC variability in models and observations

Abstract Interconnections between ocean basins are recognized as an important driver of climate variability. Recent modeling evidence suggests that the North Atlantic climate can respond to persistent warming of the tropical Indian Ocean sea surface temperature (SST) relative to the rest of the tropics (rTIO). Here, we use observational data to demonstrate that multi-decadal changes in pantropical ocean temperature gradients lead to variations of an SST-based proxy of the Atlantic Meridional Overturning Circulation (AMOC). The largest contribution to this temperature gradient-AMOCconnection comes from gradients between the Indian and Atlantic Oceans. TherTIOindex yields the strongest connection of this tropical temperature gradient to theAMOC. Focusing on the internally generated signal in three observational products reveals that an SST-basedAMOCproxy index has closely followed low-frequency changes ofrTIOtemperature with about 26-year lag since 1870. Analyzing the pre-industrial control simulations of 44 CMIP6 climate models shows that theAMOCproxy index lags simulated mid-latitudeAMOCvariations by 4 ± 4 years. These model simulations reveal the mechanism connectingAMOCvariations to pantropical ocean temperature gradients at a 27 ± 2 years lag, matching the observed time lag in 28 out of the 44 analyzed models. rTIO temperature changes affect the North Atlantic climate through atmospheric planetary waves, impacting temperature and salinity in the subpolar North Atlantic, which modifies deep convection and ultimately the AMOC. Through this mechanism, observed internalrTIOvariations can serve as a multi-decadal precursor ofAMOCchanges with important implications forAMOCdynamics and predictability.

Meteorology & Atmospheric Sciences↗

DOE OSTI2020

Microparticle cloud imaging and tracking for data-driven plasma science

Oceans of image and particle track data encountered in plasma interactions with microparticle clouds motivate development and applications of machine-learning (ML) algorithms. A local-constant-velocity tracker, a Kohonen neural network or self-organizing map, the feature tracking kit, and U-Net are described and compared with each other for microparticle cloud datasets generated from exploding wires, dusty plasmas, and atmospheric plasmas. Particle density and the signal-to-noise ratio have been identified as two important factors that affect the tracking accuracy. Fast Fourier transform is used to reveal how U-Net, a deep convolutional neural network developed for non-plasma applications, achieves the improvements for noisy scenes. Viscous effects are revealed in the ballistic motions of the particles from the exploding wires and atmospheric plasmas. Subdiffusion of microparticles satisfying Δr 2 ∝t k (k=0.84±0.02) is obtained from the dusty plasma datasets. Microparticle cloud imaging and tracking, when enhanced with data and ML models, present new possibilities for plasma physics.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2025

Improved heavy-ion PID using scintillation light detector with neural network analysis: a Monte Carlo simulation study

The photon collection efficiency of gaseous scintillator detectors varies according to the position of the impinging charged particles in the medium that generates scintillation light. Thus, when impinging particles are distributed over a large area, the intrinsic photon-number resolution of the system is affected by a large variation. This work presents and discusses a method for adjusting the total number of detected photons to account for variation in the photon collection efficiency as a function of the position of the light source within the scintillating medium. The method was developed and validated by processing data from systematic simulation studies based on GEANT4 that model the response of the Energy Loss Optical Scintillation System (ELOSS) detector. The position of the charged particle is calculated using a deep neural network algorithm. This is accomplished by analyzing the distribution of scintillation light recorded by the array of photosensors. The estimated particle position is then used to calculate the correction factor and adjust the amount of captured light to account for variations in the photon collection efficiency. The neural network algorithm provides excellent tracking capabilities, achieving sub-millimeter position resolution and an angular resolution of 12 mrad, approaching the performance of traditional tracking detectors (e.g., drift chambers). The present method can be generalized to any optical scintillation system where the photon collection efficiency depends on the position of the impinging particle.

Heavy-ion detectors↗

DOE OSTI2023

MACHINE LEARNING-ENABLED PREDICTION OF TRANSIENT INJECTION MAP IN AUTOMOTIVE INJECTORS WITH UNCERTAINTY QUANTIFICATION

Accurate prediction of injection profiles is a critical aspect of linking injector operation with engine performance and emissions. However, highly resolved injector simulations can take one to two weeks of wall-clock time, which is incompatible with engine design cycles with desired turnaround times of less than a day. Hence, it is important to reduce the time-to-solution of the internal flow simulations by several orders of magnitude to make it compatible with engine simulations. This work demonstrates a data-driven approach for tackling the computational overhead of injector simulations, whereby the transient injection profiles are emulated for a side-oriented, single-hole diesel injector using a Bayesian machine-learning framework. First, an interpretable Bayesian learning strategy was employed to understand the effect of design parameters on the total void fraction field. Then, autoencoders are utilized for efficient dimensionality reduction of the flowfields. Gaussian process models are finally used to predict the spatiotemporal void fraction field at the injector exit for unknown operating conditions. The Gaussian process models produce principled uncertainty estimates associated with the emulated flowfields, which provide the engine designer with valuable information of where the data-driven predictions can be trusted in the design space. The Bayesian flowfield predictions are compared with the corresponding predictions from a deep neural network, which has been transfer-learned from static needle simulations from a previous work by the authors. The emulation framework can predict the void fraction field at the exit of the orifice within a few seconds, thus achieving a speed-up factor of up to 38 x 10(6) over the traditional simulation-based approach of generating transient injection maps.

machine learning↗