Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “network embeddings”

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 253 records · Page 14
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Smart Pixels: In-pixel AI for on-sensor data filtering

We present a smart pixel prototype readout integrated circuit (ROIC) designed in CMOS 28 nm bulk process, with in-pixel implementation of an artificial intelligence (AI) / machine learning (ML) based data filtering algorithm designed as proof-of-principle for a Phase III upgrade at the Large Hadron Collider (LHC) pixel detector. The first version of the ROIC consists of two matrices of 256 smart pixels, each 25$\times$25 µm\textsuperscript{2} in size. Each pixel consists of a charge-sensitive preamplifier with leakage current compensation and three auto-zero comparators for a 2-bit flash-type ADC. The frontend is capable of synchronously digitizing the sensor charge within 25 ns. Measurement results show an equivalent noise charge (ENC) of $\sim$30e\textsuperscript{-} and a total dispersion of $\sim$100e\textsuperscript{-} The second version of the ROIC uses a fully connected two-layer neural network (NN) to process information from a cluster of 256 pixels to determine if the pattern corresponds to highly desirable high-momentum particle tracks for selection and readout. The digital NN is embedded in-between analog signal processing regions of the 256 pixels without increasing the pixel size and is implemented as fully combinatorial digital logic to minimize power consumption and eliminate clock distribution, and is active only in the presence of an input signal. The total power consumption of the neural network is $\sim$ 300 $\mu$W. The NN performs momentum classification based on the generated cluster patterns and even with a modest momentum threshold, it is capable of 54.4\% – 75.4\% total data rejection, opening the possibility of using the pixel information at 40MHz for the trigger. The total power consumption of analog and digital functions per pixel is $\sim$ 6 $\mu$W per pixel, which corresponds to $\sim$ 1 W/cm\textsuperscript{2} staying within the experimental constraints.

Parpillon, Benjamin↗
DOE OSTI2024

Smart Pixels: In-pixel AI for on-sensor data filtering

We present a smart pixel prototype readout integrated circuit (ROIC) designed in CMOS 28 nm bulk process, with in-pixel implementation of an artificial intelligence (AI) / machine learning (ML) based data filtering algorithm designed as proof-of-principle for a Phase III upgrade at the Large Hadron Collider (LHC) pixel detector. The first version of the ROIC consists of two matrices of 256 smart pixels, each 25$\times$25 $\mu$m$^2$ in size. Each pixel consists of a charge-sensitive preamplifier with leakage current compensation and three auto-zero comparators for a 2-bit flash-type ADC. The frontend is capable of synchronously digitizing the sensor charge within 25 ns. Measurement results show an equivalent noise charge (ENC) of $\sim$30e$^-$ and a total dispersion of $\sim$100e$^-$ The second version of the ROIC uses a fully connected two-layer neural network (NN) to process information from a cluster of 256 pixels to determine if the pattern corresponds to highly desirable high-momentum particle tracks for selection and readout. The digital NN is embedded in-between analog signal processing regions of the 256 pixels without increasing the pixel size and is implemented as fully combinatorial digital logic to minimize power consumption and eliminate clock distribution, and is active only in the presence of an input signal. The total power consumption of the neural network is $\sim$ 300 $\mu$W. The NN performs momentum classification based on the generated cluster patterns and even with a modest momentum threshold, it is capable of 54.4% - 75.4% total data rejection, opening the possibility of using the pixel information at 40MHz for the trigger. The total power consumption of analog and digital functions per pixel is $\sim$ 6 $\mu$W per pixel, which corresponds to $\sim$ 1 W/cm$^2$ staying within the experimental constraints.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2024

SPARTAN (Scalable Probabilistic Application Reconfigurable Tensor Autonomous Network)

The technical founder of Ludwig Computing Inc has been competitively selected for support by Cyclotron Road, a U.S. Department of Energy (DOE) Advanced Manufacturing Office (AMO) Lab-Embedded Entrepreneurship Program (LEEP) through an approved merit review process. Ludwig Computing Inc, supported by the U.S. Department of Energy's Advanced Manufacturing Office through the Cyclotron Road program, has investigated the advantages of probabilistic computing for real-world compute-intensive applications. This research adds to the understanding of alternative computing paradigms by exploring a unique hardware-software co-design that integrates quantum computing methods with nature-inspired problem-solving techniques. The project's focus on areas such as combinatorial optimization, graph analytics, and machine learning demonstrates the potential for significant advancements in computational efficiency and performance. By harnessing natural randomness to streamline large circuits into fewer devices, Ludwig's approach enables massive parallelism, potentially offering higher throughput, speed, and energy efficiency compared to conventional hardware solutions. This work benefits the public by paving the way for more efficient computing solutions that could address complex real-world problems while potentially reducing energy consumption in data-intensive industries.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Latent code-based fusion: A Volterra neural network approach

We propose a deep structure encoder using Volterra Neural Networks (VNNs) to seek a latent representation of multi-modal data whose features are jointly captured by a union of subspaces. The so-called self-representation embedding of the latent codes leads to a simplified fusion which is driven by a similarly constructed decoding. The Volterra Filter architecture achieved reduction in parameter complexity is primarily due to controlled non-linearities being introduced by the higher-order convolutions in lieu of generalized activation functions. Experimental results on two different datasets have shown a significant improvement in the clustering performance for VNNs auto-encoder over conventional Convolutional Neural Networks (CNNs) auto-encoder. In addition, we also show that the proposed approach demonstrates a much-improved sample complexity over CNN-based auto-encoder with a robust classification performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Parameter Estimation for Decoding Sensor Signals

This paper introduces a parameter estimation approach for decoding digital sensor signals in a cyber-physical system. For unknown or not fully characterized digital sensor data, it can be difficult to decipher a desired signal from background or noise. In a cyber-physical system with networked sensors, we can leverage knowledge of the physical system to inform the decoding of the digital signals. This work in progress is a case study on deciphering commercial vehicle on-board sensor networks that communicate through the Controller Area Network (CAN). By understanding the stock vehicle sensor network, a vehicle can be extended into a scalable research platform with minimal instrumentation. Our challenge was to localize desired sensor signals encoded in network traffic that included other sensor data, control messages, as well as encoding and security overhead. Due to the vehicle’s unknown sensor network, our approach developed methods to efficiently analyze and identify key signals despite the large state-space for potential signal embeddings.

Nice, Matthew↗
DOE OSTI2023

GRIDS-Net: Inverse shape design and identification of scatterers via geometric regularization and physics-embedded deep learning

This study presents a deep learning based methodology for both remote sensing and design of acoustic scatterers. The ability to determine the shape of a scatterer, either in the context of material design or sensing, plays a critical role in many practical engineering problems. This class of inverse problems is extremely challenging due to their high-dimensional, nonlinear, and ill-posed nature. To overcome these technical hurdles, we introduce a geometric regularization approach for deep neural networks (DNN) based on non-uniform rational B-splines (NURBS) and capable of predicting complex 2D scatterer geometries in a parsimonious dimensional representation. Then, this geometric regularization is combined with physics-embedded learning and integrated within a robust convolutional autoencoder (CAE) architecture to accurately predict the shape of 2D scatterers in the context of identification and inverse design problems. Further, an extensive numerical study is presented in order to showcase the remarkable ability of this approach to handle complex scatterer geometries while generating physically-consistent acoustic fields. The study also assesses and contrasts the role played by the (weakly) embedded physics in the convergence of the DNN predictions to a physically consistent inverse design.

42 ENGINEERING↗
DOE OSTI2025

Knowledge Oriented Graph Unified Transformer (KOGUT) v0.1

KOGUT — Knowledge Oriented Graph Unified Transformer KOGUT implements the Relational Graph Transformer (RelGT) architecture for knowledge graph link prediction in biological domains, with a primary focus on microbial growth media prediction. While the original RelGT (arXiv:2505.10960) targets relational tables, time series, and multi-table databases, KOGUT adapts this architecture for heterogeneous biological knowledge graphs, providing first-in-class AI predictive models for microbial cultivation. Key Adaptations Beyond Original RelGT: - Knowledge Graph Focus: Applied to biological KGs with semantic node types (taxa, chemicals, media, phenotypes, environments) versus generic relational database tables, trained on the KG-Microbe knowledge graph (1.3M entities, 2.9M edges, 24 relation types). - Multimodal Node Encoding: Integrates node labels, categories, descriptions, and synonyms from KG metadata through learned embedding layers—adapting relational column features to graph node attributes with textual semantics. - Extended K-Hop Subgraph Strategy: Optimized neighborhood sampling (3-hop default, configurable up to 200 nodes) tuned for sparse biological networks, building on the original local-global attention framework with biological relation preservation. - Biolink Predicate Preservation: Type-specific transformations for 24 biological edge semantics (occurs_in, consumes, produces, has_phenotype, subclass_of) beyond standard relational foreign keys, enabling multi-relation link prediction. - Inductive Learning Support: Enables zero-shot predictions for novel taxa through feature-based embeddings (temperature, oxygen requirements, gram stain, cell shape), extending the original transductive relational benchmark scope to uncultured microorganisms. CheapSOTA Performance Optimizations (This Distribution): - VQ-EMA Centroid Attention: Vector quantization with exponential moving average for improved global context modeling (+5-10% MRR improvement). - HDF5 Precomputed Data Loading: One-time preprocessing of k-hop subgraphs to eliminate redundant graph traversals (2-5× training speedup). - Distributed Data Parallel Training: Multi-GPU support for scaling to larger knowledge graphs (tested on 4× NVIDIA A100 GPUs at NERSC Perlmutter). - Mixed Precision Training: Automatic mixed precision (AMP) for memory efficiency and faster training. Advantages Over Standard Knowledge Graph Embedding Models: Combines RelGT's proven multi-element tokenization (features, type, hop, structure) with graph-native biological representations, enabling interpretable link prediction across heterogeneous entities that standard embedding models (TransE, RotatE, ComplEx) and table-based transformers cannot directly model. Achieves near-perfect performance on microbial growth media prediction (MRR: 0.9966, Precision@1: 0.9932, Hit@10: 1.0000) while maintaining explainability through attention-based reasoning over biological pathways. Training Data: - KG-Microbe merged knowledge graph: 1,379,337 nodes, 2,960,472 edges - 24 biological relation types including taxonomic hierarchies, metabolic interactions, phenotype associations, and environmental relationships - Primary prediction task: Growth media suitability for microbial taxa (biolink:occurs_in, 50K edges) - Multi-relation capability: Predicts links for any of the 24 relation types, including chemical consumption/production, phenotype associations, and taxonomic classification Citation: Original RelGT Architecture: Dwivedi et al., "Relational Graph Transformer", arXiv:2505.10960, 2025 KOGUT Implementation: Knowledge Oriented Graph Unified Transformer for Microbial Growth Media Prediction Developed at Lawrence Berkeley National Laboratory (LBNL) Trained on NERSC Perlmutter supercomputer

Joachimiak, Marcin [Lawrence Berkeley National Lab↗
DOE OSTI2021

Secure LoRa Firmware Update with Adaptive Data Rate Techniques

Internet of Things (IoT) devices rely upon remote firmware updates to fix bugs, update embedded algorithms, and make security enhancements. Remote firmware updates are a significant burden to wireless IoT devices that operate using low-power wide-area network (LPWAN) technologies due to slow data rates. One LPWAN technology, Long Range (LoRa), has the ability to increase the data rate at the expense of range and noise immunity. The optimization of communications for maximum speed is known as adaptive data rate (ADR) techniques, which can be applied to accelerate the firmware update process for any LoRa-enabled IoT device. In this paper, we investigate ADR techniques in an application that provides remote monitoring of cattle using small, battery-powered devices that transmit data on cattle location and health using LoRa. In addition to issues related to firmware update speed, there are significant concerns regarding reliability and security when updating firmware on mobile, energy-constrained devices. A malicious actor could attempt to steal the firmware to gain access to embedded algorithms or enable faulty behavior by injecting their own code into the device. A firmware update could be subverted due to cattle moving out of the LPWAN range or the device battery not being sufficiently charged to complete the update process. To address these concerns, we propose a secure and reliable firmware update process using ADR techniques that is applicable to any mobile or energy-constrained LoRa device. The proposed system is simulated and then implemented to evaluate its performance and security properties.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Materials Genomics Search for Possible Helium‐Absorbing Nano‐Phases in Fusion Structural Materials

Abstract Civilian fusion demands structural materials that can withstand the harsh environments imposed inside fusion plasma reactors. The structural materials often transmute under 14.1 MeV fast neutrons, producing helium (He), which embrittles the grain boundary (GB) network. Here, it is shown that neutron‐friendly and mechanically strong nano‐phases with atomic‐scale free volume can have low He‐embedding energy and >10 at.% He‐absorbing capacity, and can be especially advantageous for soaking up He on top of resisting radiation damage and creep, provided they have thermodynamic compatibility with the matrix phase, satisfactory equilibrium wetting angle, as well as a high enough melting point. The preliminary experimental demonstration proves that is a good ab initio predictor of He shielding potency in nano‐heterophase materials, and thus, is used as a key feature for computational screening. In this context, a list of viable compounds expected to be good He‐absorbing nano‐phases is presented, taking into account , the neutron absorption and activation cross‐sections, the elastic moduli, melting temperature, the thermodynamic compatibility, and the equilbrium wetting angle of the nano‐phases with the Fe matrix as an example.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Moisture Absorbing and Water Self‐Releasing from Hybrid Hydrogel Desiccants

Abstract Atmospheric moisture is a valuable resource for fresh water and potentially sustainable energy. However, direct harvesting water from moisture (the vapor form) remains the most challenging. Hybrid desiccants made from hydrogels embedded with salt offer promise in absorbing moisture. However, the desorption process often requires additional energy to heat the samples. Here, poly(acrylic acid) (PAA) hydrogels embedded with lithium chloride are prepared, demonstratng simultaneous moisture absorption and self‐release of liquid water at room temperature with 50–90% relative humidity. The water self‐releasing process can be separated into two distinct stages: 1) surface release, where water droplets grow on the hydrogel surface due to differences in nucleation and diffusion rates, and 2) bulk release, triggered by the collapse of polymer chains, subsequently releasing water from the hydrogel network. Factors such as salt concentration, hydrogel crosslinking density, and film thickness are investigated to better understand the moisture absorption and water‐releasing processes. Moreover, hydrophobic domains are introduced onto the salt‐embedded PAA films, creating an edge effect that enhances the droplet growth rates. When the hydrophobic domains are patterned, the movement of released water can be guided, resulting in a threefold increase in water removal rate attributed to gravitational force.

Chemistry↗
DOE OSTI2022

Modeling design and control problems involving neural network surrogates

Here, we consider nonlinear optimization problems that involve surrogate models represented by neural networks. We demonstrate first how to directly embed neural network evaluation into optimization models, highlight a difficulty with this approach that can prevent convergence, and then characterize stationarity of such models. We then present two alternative formulations of these problems in the specific case of feedforward neural networks with ReLU activation: as a mixed-integer optimization problem and as a mathematical program with complementarity constraints. For the latter formulation we prove that stationarity at a point for this problem corresponds to stationarity of the embedded formulation. Each of these formulations may be solved with state-of-the-art optimization methods, and we show how to obtain good initial feasible solutions for these methods. We compare our formulations on three practical applications arising in the design and control of combustion engines, in the generation of adversarial attacks on classifier networks, and in the determination of optimal flows in an oil well network.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Decomposition Algorithms for Solving NP-hard Problems on a Quantum Annealer

NP-hard problems such as the maximum clique or minimum vertex cover problems, two of Karp’s 21 NP-hard problems, have several applications in computational chemistry, biochemistry and computer network security. Adiabatic quantum annealers can search for the optimum value of such NP-hard optimization problems, given the problem can be embedded on their hardware. However, this is often not possible due to certain limitations of the hardware connectivity structure of the annealer. This paper studies a general framework for a decomposition algorithm for NP-hard graph problems aiming to identify an optimal set of vertices. Our generic algorithm allows us to recursively divide an instance until the generated subproblems can be embedded on the quantum annealer hardware and subsequently solved. Furthermore, the framework is applied to the maximum clique and minimum vertex cover problems, and we propose several pruning and reduction techniques to speed up the recursive decomposition. The performance of both algorithms is assessed in a detailed simulation study.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Elucidating interprotein energy transfer dynamics within the antenna network from purple bacteria

In photosynthesis, absorbed light energy transfers through a network of antenna proteins with near-unity quantum efficiency to reach the reaction center, which initiates the downstream biochemical reactions. While the energy transfer dynamics within individual antenna proteins have been extensively studied over the past decades, the dynamics between the proteins are poorly understood due to the heterogeneous organization of the network. Previously reported timescales averaged over such heterogeneity, obscuring individual interprotein energy transfer steps. Here, we isolated and interrogated interprotein energy transfer by embedding two variants of the primary antenna protein from purple bacteria, light-harvesting complex 2 (LH2), together into a near-native membrane disc, known as a nanodisc. We integrated ultrafast transient absorption spectroscopy, quantum dynamics simulations, and cryogenic electron microscopy to determine interprotein energy transfer timescales. By varying the diameter of the nanodiscs, we replicated a range of distances between the proteins. The closest distance possible between neighboring LH2, which is the most common in native membranes, is 25 Å and resulted in a timescale of 5.7 ps. Larger distances of 28 to 31 Å resulted in timescales of 10 to 14 ps. Corresponding simulations showed that the fast energy transfer steps between closely spaced LH2 increase transport distances by ∼15%. Overall, our results introduce a framework for well-controlled studies of interprotein energy transfer dynamics and suggest that protein pairs serve as the primary pathway for the efficient transport of solar energy.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

FPGA-accelerated SpeckleNN with SNL for real-time X-ray single-particle imaging

We present the implementation of a specialized version of our previously published unified embedding model, SpeckleNN, for real-time speckle pattern classification in X-ray Single-Particle Imaging (SPI), using the SLAC Neural Network Library (SNL) on an FPGA platform. This hardware realization transitions SpeckleNN from a prototypic model into a practical edge solution, optimized for running inference near the detector in high-throughput X-ray free-electron laser (XFEL) facilities, such as those found at the Linac Coherent Light Source (LCLS). To address the resource constraints inherent in FPGAs, we developed a more specialized version of SpeckleNN. The original model, which was designed for broader classification across multiple biological samples, comprised ~5.6 million parameters. The new implementation, while reducing the parameter count to 64.6K (a 98.8% reduction), focuses on maintaining the model's essential functionality for real-time operation, achieving an accuracy of 90%. Furthermore, we compressed the latent space from 128 to 50 dimensions. This implementation was demonstrated on the KCU1500 FPGA board, utilizing 71% of available DSPs, 75% of LUTs, and 48% of FFs, with an average power consumption of 9.4W according to the Vivado post-implementation report. The FPGA performed inference on a single image with a latency of 45.015 microseconds at a 200 MHz clock rate. In comparison, running the same inference on an NVIDIA A100 GPU resulted in an average power consumption of ~73W and an image processing latency of around 400 microseconds. Our FPGA-accelerated version of SpeckleNN demonstrated significant improvements, achieving an 8.9 × speedup and a 7.8 × reduction in power consumption compared to the GPU implementation. Key advancements include model specialization and dynamic weight loading through SNL, which eliminates the need for time-consuming FPGA design re-synthesis, allowing fast and continuous deployment of models (re)trained online. These innovations enable real-time adaptive classification and efficient vetoing of speckle patterns, making SpeckleNN more suited for deployment in XFEL facilities. This implementation has the potential to significantly accelerate SPI experiments and enhance adaptability to evolving experimental conditions.

47 OTHER INSTRUMENTATION↗
DOE OSTI2021

Physics-informed machine learning of the Lagrangian dynamics of velocity gradient tensor

Reduced models describing the Lagrangian dynamics of the velocity gradient tensor (VGT) in homogeneous isotropic turbulence (HIT) are developed under the physics-informed machine learning (PIML) framework. We consider the VGT at both Kolmogorov scale and coarse-grained scale within the inertial range of HIT. Building reduced models requires resolving the pressure Hessian and subfilter contributions, which is accomplished by constructing them using the integrity bases and invariants of the VGT. The developed models can be expressed using the extended tensor basis neural network (TBNN) introduced by Ling et al. [J. Fluid Mech. 807, 155 (2016)]. Physical constraints, such as Galilean invariance, rotational invariance, and incompressibility condition, are thus embedded in the models explicitly. Our PIML models are trained on the Lagrangian data from a high-Reynolds number direct numerical simulation (DNS). To validate the results, we perform a comprehensive out-of-sample test. We observe that the PIML model provides an improved representation for the magnitude and orientation of the small-scale pressure Hessian contributions. Statistics of the flow, as indicated by the joint PDF of second and third invariants of the VGT, show good agreement with the “ground-truth” DNS data. A number of other important features describing the structure of HIT are reproduced by the model successfully. We have also identified challenges in modeling inertial range dynamics, which indicates that a richer modeling strategy is required. This helps us identify important directions for future research, in particular towards including inertial range geometry into the TBNN.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2020

Mapping the Temperature-dependent and network site-specific onset of spectral diffusion at the surface of a water cluster cage

We explore the kinetic processes that sustain equilibrium in a microscopic, finite system. This is accomplished by monitoring the spontaneous, time-dependent frequency evolution (the frequency autocorrelation) of a single OH oscillator, embedded in a water cluster held in a temperature-controlled ion trap. The measurements are carried out by applying two-color, IR-IR photodissociation mass spectrometry to the D3O+?(HDO)(D2O)19 isotopologue of the “magic number” protonated water cluster, H+?(H2O)21. The OH group can occupy any one of the five spectroscopically distinct sites in the distorted pentagonal dodecahedron cage structure. The OH frequency is observed to evolve over tens of milliseconds in the temperature range (90-120 K). Starting at 100 K, large “jumps” are observed between two OH frequencies separated by ~300 cm-1 indicating migration of the OH group from the bound OH site at 3350 cm-1 to the free position at 3686 cm-1. Increasing the temperature to 110 K leads to partial interconversion among many sites. All sites are observed to interconvert at 120 K such that the distribution of the unique OH group among them adopts the form one would expect for a canonical ensemble. The spectral dynamics displayed by the clusters thus offer an unprecedented view into the molecular-level processes that drive spectral diffusion in an extended network of water molecules.

Yang, Nan↗
DOE OSTI2020

Application of an Embedded Fracture and Borehole Modeling Approach to the Understanding of EGS Collab Experiment 1

EGS Collab is a series of meso-scale experiments and associated numerical simulation activities being funded by the United States Department of Energy, Geothermal Technologies Office (GTO) to investigate enhanced geothermal system processes under in-situ stress and slightly elevated temperature conditions. This project is designed to provide scientists and engineers with immediate access to impermeable rock at scales larger than possible in the laboratory, but generally smaller than those for commercial production. Immediate access to rock is provided via the existing drifts of the former Homestake Gold Mine, now operated as the Sanford Underground Research Facility in Lead, South Dakota. The objectives of the EGS Collab project are to develop well controlled fracture networks between injector and producer boreholes using normal and shear stimulation for permeability enhancement. The first experimental site was located off the West Access drift on the 4850 Level (4850 feet below ground surface) in phyllite of the Precambrian Poorman formation, and involved the creation of a fracture network comprising a combination of hydraulic and natural fractures. A second experimental site is now being considered near the battery alcove on the 4100 Level in amphibolite and rhyolite of the Yates Unit. Data generated during these experiments will be compared against predictions of a suite of computer codes specifically designed to solve problems involving coupled thermal, hydrological, geomechanical, and geochemical processes. Comparisons between experimental and numerical simulation results will provide code developers with direction for improvements and verification of process models, build confidence in the suite of available numerical tools, and ultimately identify critical future development needs for the geothermal modeling community. Moreover, conducting thorough comparisons of models, modelling approaches, measurement approaches and measured data, via the EGS Collab project, will serve to identify techniques that are most likely to succeed at the Frontier Observatory for Research in Geothermal Energy (FORGE), the GTO’s flagship EGS research effort. Experiment 1 has comprised a series of successful tests, including a long-term chilled-water circulation test, but the testbed has two atypical EGS elements. Active ventilation in the adjacent drift over a 50-year period cooled the testbed rock mass and hydraulic fracturing intersected monitoring boreholes making them conduits for fluid flow. Numerical simulations executed in support of the design of the EGS Collab Experiment 1, computed a radial temperature distribution and stress gradient orthogonal to the drift axis, resulting in the forecast of an oblong hydraulic fracture geometry, extended in the direction of the drift from the stimulation borehole. This paper describes the numerical simulation of long-term chilled-water circulation test with an embedded fracture and borehole modeling approach. The principal objective of the simulation work is show agreement between experimental observations in terms of production fluid temperatures, monitoring borehole temperatures, injection and production pressures, and tracer recoveries using a single conceptual model for the fracture network, using characterization and monitoring data generally available for EGS. A secondary objective is to improve the numerical simulation result comparisons with additional data available from the broader monitoring equipment within the testbed versus deeper, hotter, and more remote EGS. One complicating factor for comparisons between the numerical simulation results and experimental observations, is the Joule-Thomson heating associated with large pressure drops across the fracture network.

numerical simulation, embedded fractures and boreh↗
DOE OSTI2021

Reduced-order modeling of advection-dominated systems with recurrent neural networks and convolutional autoencoders

A common strategy for the dimensionality reduction of nonlinear partial differential equations (PDEs) relies on the use of the proper orthogonal decomposition (POD) to identify a reduced subspace and the Galerkin projection for evolving dynamics in this reduced space. However, advection-dominated PDEs are represented poorly by this methodology since the process of truncation discards important interactions between higher-order modes during time evolution. In this study, we demonstrate that encoding using convolutional autoencoders (CAEs) followed by a reduced-space time evolution by recurrent neural networks overcomes this limitation effectively. We demonstrate that a truncated system of only two latent space dimensions can reproduce a sharp advecting shock profile for the viscous Burgers equation with very low viscosities, and a six-dimensional latent space can recreate the evolution of the inviscid shallow water equations. Additionally, the proposed framework is extended to a parametric reduced-order model by directly embedding parametric information into the latent space to detect trends in system evolution. Furthermore, our results show that these advection-dominated systems are more amenable to low-dimensional encoding and time evolution by a CAE and recurrent neural network combination than the POD-Galerkin technique.

97 MATHEMATICS AND COMPUTING↗