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 181 records · Page 10

All sources NASA NTRS DOE OSTI

DOE OSTI2021

Integrating deep neural networks with full-waveform inversion: Reparameterization, regularization, and uncertainty quantification

Full-waveform inversion (FWI) is an accurate imaging approach for modeling the velocity structure by minimizing the misfit between recorded and predicted seismic waveforms. However, the strong nonlinearity of FWI resulting from fitting oscillatory waveforms can trap the optimization in local minima. We have adopted a neural-network-based full-waveform inversion (NNFWI) method that integrates deep neural networks with FWI by representing the velocity model with a generative neural network. Neural networks can naturally introduce spatial correlations as regularization to the generated velocity model, which suppresses noise in the gradients and mitigates local minima. Furthermore, the velocity model generated by neural networks is input to the same partial differential equation (PDE) solvers used in conventional FWI. The gradients of the neural networks and PDEs are calculated using automatic differentiation, which back propagates gradients through the acoustic PDEs and neural network layers to update the weights of the generative neural network. Experiments on 1D velocity models, the Marmousi model, and the 2004 BP model determine that NNFWI can mitigate local minima, especially for imaging high-contrast features such as salt bodies, and it significantly improves the inversion in the presence of noise. Adding dropout layers to the neural network model also allows analyzing the uncertainty of the inversion results through Monte Carlo dropout. NNFWI opens a new pathway to combine deep learning and FWI for exploiting the characteristics of deep neural networks and the high accuracy of PDE solvers. Because NNFWI does not require extra training data and optimization loops, it provides an attractive and straightforward alternative to conventional FWI.

58 GEOSCIENCES↗

DOE OSTI2024

Differentiable modeling and optimization of non-aqueous Li-based battery electrolyte solutions using geometric deep learning

Electrolytes play a critical role in designing next-generation battery systems, by allowing efficient ion transfer, preventing charge transfer, and stabilizing electrode-electrolyte interfaces. In this work, we develop a differentiable geometric deep learning (GDL) model for chemical mixtures, DiffMix, which is applied in guiding robotic experimentation and optimization towards fast charging battery electrolytes. In particular, we extend mixture thermodynamic and transport laws by creating GDL-learnable physical coefficients. We evaluate our model with mixture thermodynamics and ion transport properties, where we show improved prediction accuracy and model robustness of Diff-Mix than its purely data-driven variants. Furthermore, with a robotic experimentation setup, Clio, we improve ionic conductivity of electrolytes by over 18.8% within 10 experimental steps, via differentiable optimization built on DiffMix gradients. By combining GDL, mixture physics laws, and robotic experimentation, DiffMix expands the predictive modeling methods for chemical mixtures and enables efficient optimization in large chemical spaces.

25 - ENERGY STORAGE↗

DOE OSTI2023

Hybrid Attack Graph Generation with Graph Convolutional Deep-Q Learning

Critical infrastructures such as power grids have become increasingly complex, connected, and vulnerable to adverse scenarios, including cyber and physical attacks and faults. Effective risk mitigation for such cyber-physical energy systems (CPES), requires preemptive knowledge of likely adversarial attack scenarios. Hybrid Attack Graph (HAG) is a structured way to represent an adversarial scenario as an attack sequence using a threat model. However, the scarcity of documented attack sequences hinders analysts and CPES planners’ ability to identify credible attack scenarios for a given CPES. We propose a data-driven Graph Convolutional Deep-Q Network (GCDQ) to address this data challenge through generating HAGs. By leveraging limited real-world observations from the MITRE ATT&CK knowledge base, our GCDQ model synthesizes realistic graphs with the targeted attribute of minimum detectability via reinforcement learning. This generative model is the first step in creating a tool to substantially boost the attack sequence dataset and enhance the performance of CPS defense-related tasks by providing insights into likely attack sequences with given attributes.

deep learning, artificial intelligence↗

DOE OSTI2021

Predictive Data-driven Platform for Subsurface Energy Production

Subsurface energy activities such as unconventional resource recovery, enhanced geothermal energy systems, and geologic carbon storage require fast and reliable methods to account for complex, multiphysical processes in heterogeneous fractured and porous media. Although reservoir simulation is considered the industry standard for simulating these subsurface systems with injection and/or extraction operations, reservoir simulation requires spatio-temporal “Big Data” into the simulation model, which is typically a major challenge during model development and computational phase. In this work, we developed and applied various deep neural network-based approaches to (1) process multiscale image segmentation, (2) generate ensemble members of drainage networks, flow channels, and porous media using deep convolutional generative adversarial network, (3) construct multiple hybrid neural networks such as convolutional LSTM and convolutional neural network-LSTM to develop fast and accurate reduced order models for shale gas extraction, and (4) physics-informed neural network and deep Q-learning for flow and energy production. We hypothesized that physicsbased machine learning/deep learning can overcome the shortcomings of traditional machine learning methods where data-driven models have faltered beyond the data and physical conditions used for training and validation. We improved and developed novel approaches to demonstrate that physics-based ML can allow us to incorporate physical constraints (e.g., scientific domain knowledge) into ML framework. Outcomes of this project will be readily applicable for many energy and national security problems that are particularly defined by multiscale features and network systems.

58 GEOSCIENCES↗

DOE OSTI2023

Deep learning methods for drug response prediction in cancer: Predominant and emerging trends

Cancer claims millions of lives yearly worldwide. While many therapies have been made available in recent years, by in large cancer remains unsolved. Exploiting computational predictive models to study and treat cancer holds great promise in improving drug development and personalized design of treatment plans, ultimately suppressing tumors, alleviating suffering, and prolonging lives of patients. A wave of recent papers demonstrates promising results in predicting cancer response to drug treatments while utilizing deep learning methods. These papers investigate diverse data representations, neural network architectures, learning methodologies, and evaluations schemes. However, deciphering promising predominant and emerging trends is difficult due to the variety of explored methods and lack of standardized framework for comparing drug response prediction models. To obtain a comprehensive landscape of deep learning methods, we conducted an extensive search and analysis of deep learning models that predict the response to single drug treatments. A total of 61 deep learning-based models have been curated, and summary plots were generated. Based on the analysis, observable patterns and prevalence of methods have been revealed. This review allows to better understand the current state of the field and identify major challenges and promising solution paths.

60 APPLIED LIFE SCIENCES↗

NASA NTRS1999

Diurnal Forcing of Planetary Atmospheres

This project investigated and developed models which will assist in developing an understanding of the diurnal forcings in the atmospheres of Mars, Venus, Earth and Jupiter. The differences in the temperatures during the day, and night, generate very deep diurnal boundary layers and strong slope-related winds. A general circulation model (GCM) was adapted for computation of Martian atmospheric dynamics, to simulate the global transport of these atmospheric tracers. The model was applied to the study of the Martian seasonal water cycle. The principles have been applied to venus. The region of study is the upper cloud level, and above rather than the area close to the surface. The model was also applied in efforts to study the evolution of volcanic clouds in the Earth's stratosphere. In particular this was used to model the dispersion of sulfuric acid particles in the atmosphere. A brief study of the application of the tidal theory to the planet Jupiter has been undertaken. The goal of the study is to deduce the source of the tidal dissipation which is thought to have led to the orbital resonances amongst the Galilean satellites and considerable heating of the interiors of Io, Europa, and Ganymede.

Houben, Howard C.↗

NASA NTRS2022

A Dynamo Simulation Generating Saturn-Like Small Magnetic Dipole Tilts

Abstract Among planetary dynamos, the magnetic field of Saturn stands out in its exceptional level of axisymmetry. One of its peculiar features is that the magnetic dipole mode is tilted with respect to the planetary rotation axis by only ≈0.007° or less. Numerical dynamo simulations performed in this context have had great difficulty in producing such small dipole tilt angles without introducing ad hoc ingredients such as a latitudinally varying heat flux pattern in the outer layers or stably stratified layers (SSLs). Here we present a numerical dynamo simulation that generates a highly axisymmetric dynamo with a dipole tilt of about ≈0.0008° on average. The model consists of a deep dynamo layer and an overlying low-conductivity layer but without any SSLs. We highlight a novel mechanism where strong differential rotation generated in the atmospheric layer penetrates into the dynamo region, helping to maintain a very small magnetic dipole tilt. Plain Language Summary Saturn's dipole-dominant magnetic field exhibits a very peculiar feature: the dipole component of the planetary magnetic field is tilted by less than ≈0.007° with respect to the planetary spin axis. Numerical simulations performed in this context suggest that if a spatial heat-flux variation is imposed, along with a stably stratified region, on top of an active dynamo layer, then small dipole tilt values can be realized. Here we present a model where extremely small dipole tilt values can be achieved without these ad hoc ingredients. Our simulations demonstrate that dynamo theory allows extremely small dipole tilt values in a relatively simple model configuration.

Rakesh Kumar Yadav↗

DOE OSTI2025

Leveraging interpolation models and error bounds for verifiable scientific machine learning

Effective verification and validation techniques for modern scientific machine learning workflows are challenging to devise. Statistical methods are abundant and easily deployed, but often rely on speculative assumptions about the data and methods involved. Error bounds for classical interpolation techniques can provide mathematically rigorous estimates of accuracy, but often are difficult or impractical to determine computationally. Here, in this work, we present a best-of-both-worlds approach to verifiable scientific machine learning by demonstrating that (1) multiple standard interpolation techniques have informative error bounds that can be computed or estimated efficiently; (2) comparative performance among distinct interpolants can aid in validation goals; (3) deploying interpolation methods on latent spaces generated by deep learning techniques enables some interpretability for black-box models. We present a detailed case study of our approach for predicting lift-drag ratios from airfoil images. Code developed for this work is available in a public Github repository.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Unified Wind-Wave Growth and Spectrum Functions for All Water Depths: Field Observations and Model Results

Abstract Wind-wave development is governed by the fetch- or duration-limited growth principle that is expressed as a pair of similarity functions relating the dimensionless elevation variance (wave energy) and spectral peak frequency to fetch or duration. Combining the pair of similarity functions, the fetch or duration variable can be removed to form a dimensionless function of elevation variance and spectral peak frequency, which is interpreted as the wave energy evolution with wave age. The relationship is initially developed for quasi-neural stability and quasi-steady wind forcing conditions. Further analyses show that the same fetch, duration, and wave-age similarity functions are applicable to unsteady wind forcing conditions, including rapidly accelerating and decelerating mountain gap wind episodes and tropical cyclone (TC) wind fields. Here it is shown that with the dimensionless frequency converted to dimensionless wavenumber using the surface wave dispersion relationship, the same similarity function is applicable in all water depths. Field data collected in shallow to deep waters and mild to TC wind conditions and synthetic data generated by spectrum model computations are assembled to illustrate the applicability. For the simulation work, the finite-depth wind-wave spectrum model and its shoaling function are formulated for variable spectral slopes. Given wind speed, wave age, and water depth, the measured and spectrum-computed significant wave heights and the associated growth parameters are in good agreement in forcing conditions from mild to TC winds and in all depths from deep ocean to shallow lake. Significance Statement This paper presents a growth function and spectrum model to describe wind-wave development in all water depths. Their applicability covers a wide range of wind forcing conditions including steady, accelerating, decelerating, and tropical cyclone events. Support for the unified spectrum model and growth function is presented with field observations and numerical computations.

Hwang, Paul A.↗

NASA NTRS2005

Simulator of Space Communication Networks

Multimission Advanced Communications Hybrid Environment for Test and Evaluation (MACHETE) is a suite of software tools that simulates the behaviors of communication networks to be used in space exploration, and predict the performance of established and emerging space communication protocols and services. MACHETE consists of four general software systems: (1) a system for kinematic modeling of planetary and spacecraft motions; (2) a system for characterizing the engineering impact on the bandwidth and reliability of deep-space and in-situ communication links; (3) a system for generating traffic loads and modeling of protocol behaviors and state machines; and (4) a system of user-interface for performance metric visualizations. The kinematic-modeling system makes it possible to characterize space link connectivity effects, including occultations and signal losses arising from dynamic slant-range changes and antenna radiation patterns. The link-engineering system also accounts for antenna radiation patterns and other phenomena, including modulations, data rates, coding, noise, and multipath fading. The protocol system utilizes information from the kinematic-modeling and link-engineering systems to simulate operational scenarios of space missions and evaluate overall network performance. In addition, a Communications Effect Server (CES) interface for MACHETE has been developed to facilitate hybrid simulation of space communication networks with actual flight/ground software/hardware embedded in the overall system.

Clare, Loren↗

DOE OSTI2025

Deep potential molecular dynamics simulations of low-temperature plasma-surface interactions

Machine learning approaches to potential generation for molecular dynamics (MD) simulations of low-temperature plasma-surface interactions could greatly extend the range of chemical systems that can be modeled. Empirical potentials are difficult to generalize to complex combinations of multiple elements with interactions that might include covalent, ionic, and metallic bonds. This work demonstrates that a specific machine learning approach, Deep Potential Molecular Dynamics (DeepMD), can generate potentials that provide a good model of plasma etching in the Si-Cl-Ar system. Comparisons are made between MD results using DeepMD models and empirical potentials, as well as experimental measurements. Pure Si properties predicted by the DeepMD model are in reasonable agreement with experimental results. Simulations of Si bombardment by Ar + ions demonstrate the ability of the DeepMD method to predict sputtering yields as well as the depth of the amorphous-crystalline interface. Etch yields as a function of flux ratio and ion energy for simultaneous Cl 2 and Ar + impacts are in good agreement with previous simulation results and experiment. Predictions of etch yields and etch products during plasma-assisted atomic layer etching of Si-Cl 2 -Ar are shown to be in good agreement with MD predictions using empirical potentials and with experiment. Finally, good agreement was also seen with measurements for the spontaneous etching of Si by Cl atoms at 300 K. Further, the demonstration that DeepMD can reproduce results from MD simulations using empirical potentials is a necessary condition to future efforts to extend the method to a much wider range of systems for which empirical potentials may be difficult or impossible to obtain.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2022

MARS: Malleable Actor-Critic Reinforcement Learning Scheduler

In this paper, we introduce MARS, a new scheduling system for HPC-cloud infrastructures based on a cost-aware, flexible reinforcement learning approach, which serves as an intermediate layer for next generation HPC-cloud resource manager. MARS ensembles the pre-trained models from heuristic workloads and decides on the most cost-effective strategy for optimization. A whole workflow application would be split into several optimizable dependent sub-tasks, then based on the pre- defined resource management plan, a reward will be generated after executing a scheduled task. Lastly, MARS updates the Deep Neural Network (DNN) model based on the reward. MARS is designed to optimize the existing models through reinforcement mechanisms. MARS adapts to the dynamics of workflow applications, selects the most cost-effective scheduling solution among pre-built scheduling strategies (backfilling, SJF, etc.) and self- learning deep neural network model at run-time. We evaluate MARS with different real-world workflow traces. MARS can achieve 5%-60% increased performance compare to state-of-the- art approaches.

Baheri, Betis↗

DOE OSTI2023

Recurrent Convolutional Deep Neural Networks for Modeling Time-Resolved Wildfire Spread Behavior

The increasing incidence and severity of wildfires underscores the necessity of accurately predicting their behavior. While high-fidelity models derived from first principles offer physical accuracy, they are too computationally expensive for use in real-time fire response. Low-fidelity models sacrifice some physical accuracy and generalizability via the integration of empirical measurements, but enable real-time simulations for operational use in fire response. Machine learning techniques have demonstrated the ability to bridge these objectives by learning first-principles physics while achieving computational speedups. While deep learning approaches have demonstrated the ability to predict wildfire propagation over large time periods, time-resolved fire-spread predictions are needed for active fire management. Here, in this work, we evaluate the ability of deep learning approaches in accurately modeling the time-resolved dynamics of wildfires. We use an autoregressive process in which a convolutional recurrent deep learning model makes predictions that propagate a wildfire over 15 min increments. We apply the model to four simulated datasets of increasing complexity, containing both field fires with homogeneous fuel distribution as well as real-world topologies sampled from the California region of the United States. We show that even after 100 autoregressive predictions representing more than 24 h of simulated fire spread, the resulting models generate stable and realistic propagation dynamics, achieving a Jaccard score between 0.89 and 0.94 when predicting the resulting fire scar. The inference time of the deep learning models are examined and compared, and directions for future work are discussed.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

Learning macroscopic internal variables and history dependence from microscopic models

This paper concerns the study of history dependent phenomena in heterogeneous materials in a two-scale setting where the material is specified at a fine microscopic scale of heterogeneities that is much smaller than the coarse macroscopic scale of application. Here, we specifically study a polycrystalline medium where each grain is governed by crystal plasticity while the solid is subjected to macroscopic dynamic loads. The theory of homogenization allows us to solve the macroscale problem directly with a constitutive relation that is defined implicitly by the solution of the microscale problem. However, the homogenization leads to a highly complex history dependence at the macroscale, one that can be quite different from that at the microscale. In this paper, we examine the use of machine-learning, and especially deep neural networks, to harness data generated by repeatedly solving the finer scale model to: (i) gain insights into the history dependence and the macroscopic internal variables that govern the overall response; and (ii) to create a computationally efficient surrogate of its solution operator, that can directly be used at the coarser scale with no further modeling. We do so by introducing a recurrent neural operator (RNO), and show that: (i) the architecture and the learned internal variables can provide insight into the physics of the macroscopic problem; and (ii) that the RNO can provide multiscale, specifically FE 2 , accuracy at a cost comparable to a conventional empirical constitutive relation.

36 MATERIALS SCIENCE↗

NASA NTRS2019

Deep Learning and Uncertainty Quantification for Climate Resilience

Modeling and monitoring of earth’s processes through physical models and satellite observations at high resolutions is crucial for ensuring society’s ability to adapt to climate change. Deep learning (DL) has been shown to be a valuable tool for generating high resolution data, emulating physical models, and detecting weather patterns which can then be used to inform stakeholders and decision makers. However, both the data and model parameters contain substantial uncertainties that may alter users’ decisions. In this work we present two DL applications on high-resolution climate and satellite datasets using Bayesian neural networks to generate well calibrated uncertainty estimates.

Vandal, Thomas↗

DOE OSTI2023

Texas A&M University Mobile Facility Measurements during TRACER (Field Campaign Report)

One of the main goals of the U.S Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) Tracking Aerosol Convection Interactions Experiment (TRACER) near Houston, Texas is to improve understanding of how meteorology and aerosols impact storm dynamical and microphysical processes in deep convection to better constrain and improve their model representation. The Houston area is strongly influenced by sea- and bay-breeze circulations that generate convergence and help to initiate and organize deep convection. To properly isolate and understand the roles of varied meteorological conditions and cloud condensation nuclei (CCN) and ice nucleation particles (INP) distributions in different air masses, co-located thermodynamic, kinematic, and aerosol vertical profile observations are needed. The focus of this campaign was to provide key measurements in air masses both in front of and behind sea/bay breeze fronts moving through the greater Houston area to sample the airmass heterogeneity. The overarching scientific goal of this campaign is to understand how the vertical distributions of both CCN and INP correspond to the inflow layer of deep convection in maritime, background continental, and polluted continental air masses, and how these variations influence deep convection. To fully sample the heterogeneity in both meteorological conditions and aerosols across the sea-breeze front (SBF), Texas A&M University (TAMU) deployed a InterMet 3050A 403 MHz mobile unit launching iMet-4 radiosondes and the new Rapid Onsite Atmospheric Measurement Van (ROAM-V) for aerosol sampling during the TRACER intensive operational period (IOP) from June to September 2022. The suite of instruments deployed on ROAM-V included a condensation particle counter (CPC; GRIMM Model 5.403 CPC), scanning mobility particle sizer (SMPS; TSI 3750 detector, TSI 3082 classifier, TSI 3088 neutralizer, TSI 3081A differential mobility analyzer), cloud condensation nuclei counter (Droplet Measurement Technologies CCN counter), micropulse lidar (Droplet Measurement Technologies micropulse lidar [miniMPL]), and a Davis Rotating Uniform size-cut Monitor (DRUM; DRUMAir 4-DRUM). Before sampling at each location, the latitude and longitude were recorded using the Global Positioning System (GPS) on the phone application “My Altitude”. The DRUM data were collected as part of a closely related ARM field campaign and also supported by DOE Atmospheric System Research grant DE-SC0021047. The TAMU team sampled these airmass heterogeneities by strategically choosing deployment sites in a different airmass than the ARM fixed sites in La Porte and Guy, Texas. On days when the sea/bay breeze boundary was pushing inland, the TAMU team would usually sample the airmass on the maritime side of the SBF at a coastal site in Galveston, Texas in the early afternoon (1730-1900 UTC) and then move inland ahead of the SBF to sample the airmass on the continental side during late afternoon (2030-2230). Figure 1 shows the Galveston maritime site and the array of sites for the late afternoon continental measurements.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

Cyber-Attack Detection and Accommodation for the Energy Delivery System

The goals of this project were to create a software system with a suite of key algorithms for cyber-attack detection and accommodation providing domain layer protection for critical power generation assets. Example assets included gas and steam turbines, heat recovery steam generators, and electrical generators. The aggressive algorithm goals were aimed at reducing the false positive rates in threat detection to <1% using learnings from many evolving disciplines (power turbine and generator physics, power system modeling, modern control theory, system identification, machine learning, deep learning, mathematics and data science). Additional goals for the algorithms involved localizing threats on-the-fly to know in which monitoring node the effects of attacks are present, and then providing accommodation to keep the system running uninterrupted much of the time in the presence of the attack. Accommodation had a performance goal of providing resiliency when up to 50% of monitoring nodes are in an attack state.

cybersecurity, cyber-physical↗

NASA NTRS2020

Multi-Agent Motion Planning using Deep Learning for Space Applications

State-of-the-art motion planners cannot scale to a large number of systems. Motion planning for multiple agents is an NP (non-deterministic polynomial-time) hard problem, so the computation time increases exponentially with each addition of agents. This computational demand is a major stumbling block to the motion planner's application to future NASA missions involving the swarm of space vehicles. We applied a deep neural network to transform computationally demanding mathematical motion planning problems into deep learning-based numerical problems. We showed optimal motion trajectories can be accurately replicated using deep learning-based numerical models in several 2D and 3D systems with multiple agents. The deep learning-based numerical model demonstrates superior computational efficiency with plans generated 1000 times faster than the mathematical model counterpart.

Madani, Ramtin↗