Developing efficient deep learning model for predicting copolymer properties
We report data and topology selection methods for developing an efficient deep neural network (DNN) model that predicts sequenced-defined properties of polymers.
Chemistry↗
SEARCH · Engineering Papers
Results for “Network 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.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Characterizing Reactive Transport Behavior in a Three-Dimensional Discrete Fracture Network
While several studies have linked network and in-fracture scale properties to conservative transport behavior in subsurface fractured media, studies on reactive transport cases remain relatively underdeveloped. In this study, we explore the behavior of an irreversible kinetic reaction during the interaction of two solute plumes, one consisting of species A and the other species B. When the plumes converge, these species react kinetically to form a new species C via A+B→kC. This reactive system is studied using a three-dimensional discrete fracture network (DFN) model coupled with reactive Lagrangian particle tracking. We find that the interplay of network topology and chemical properties of the reactive solutes controls reactive transport processes. The network topology drives species A and B together, and the chemical properties dictate whether and how quickly a reaction occurs. Results demonstrate that reactions are most likely to occur in high-velocity fractures that make up the network backbone. The interplay between species’ chemical properties and transport is characterized by a non-dimensional Damköhler (Da) number. We show that the spatial distribution of reactions is sensitive to Da, which subsequently influences late-time tailing behavior in outlet breakthrough time distributions. The results of this study provide initial insights into how an irreversible reaction occurs during transport in a fracture network, using a methodology that can be applied to study reactive transport in a wide range of fractured media environments and contexts.
58 GEOSCIENCES↗
Hygro-thermo-mechanical modeling of partially saturated air-entrained concrete containing dissolved salt and exposed to freeze-thaw cycles
Highlights: • A model for cyclic freezing of air-entrained concrete containing salt is proposed. • The freeze-thaw hysteresis, also including the effect of dissolved salt, is modeled. • The diffusive and convective flux of salt is coupled to the freeze-thaw processes. • The freeze-thaw response of air-entrained concrete containing salt is reproduced. • Liquid absorption from reservoirs increases with an increased concentration of salt. In cold regions, understanding the freeze-thaw behavior of air-entrained concrete is important for designing durable structures and assessing the remaining service life of existing structures. This study presents a hygro-thermo-mechanical multiphase model that describes the cyclic freeze-thaw behavior of partially saturated air-entrained concrete containing dissolved salt. An equilibrium and a non-equilibrium approach are adopted to model the ice formation, including the freeze-thaw hysteresis, inside the porous network. The model also considers the diffusive and convective transport of the dissolved salt coupled to the freeze-thaw processes. Two examples are presented to verify and highlight the capabilities of the model. The first example shows that the model is capable of reproducing the experimentally observed mechanical response of specimens containing NaC1-solutions of different concentrations. In the second example, a larger absorption of liquid from an external reservoir is obtained with an increasing salt concentration in the reservoir, which is consistent with experimental observations.
36 MATERIALS SCIENCE↗
The future of Earth system prediction: Advances in model-data fusion
Predictions of the Earth system, such as weather forecasts and climate projections, require models informed by observations at many levels. Some methods for integrating models and observations are very systematic and comprehensive (e.g., data assimilation), and some are single purpose and customized (e.g., for model validation). We review current methods and best practices for integrating models and observations. We highlight how future developments can enable advanced heterogeneous observation networks and models to improve predictions of the Earth system (including atmosphere, land surface, oceans, cryosphere, and chemistry) across scales from weather to climate. As the community pushes to develop the next generation of models and data systems, there is a need to take a more holistic, integrated, and coordinated approach to models, observations, and their uncertainties to maximize the benefit for Earth system prediction and impacts on society.
54 ENVIRONMENTAL SCIENCES↗
An investigation on machine learning predictive accuracy improvement and uncertainty reduction using VAE-based data augmentation
The confluence of ultrafast computers with large memory, rapid progress in Machine Learning (ML) algorithms, and the availability of large datasets place multiple engineering fields at the threshold of dramatic progress. However, a unique challenge in nuclear engineering is data scarcity because experimentation on nuclear systems is usually more expensive and time-consuming than most other disciplines. One potential way to resolve the data scarcity issue is deep generative learning, which uses certain ML models to learn the underlying distribution of existing data and generate synthetic samples that resemble the real data. In this way, one can significantly expand the dataset to train more accurate predictive ML models. In this study, our objective is to evaluate the effectiveness of data augmentation using variational autoencoder (VAE)-based deep generative models. We investigated whether the data augmentation leads to improved accuracy in the predictions of a deep neural network (DNN) model trained using the augmented data. Additionally, the DNN prediction uncertainties are quantified using Bayesian Neural Networks (BNN) and conformal prediction (CP) to assess the impact on predictive uncertainty reduction. To test the proposed methodology, we used TRACE simulations of steady-state void fraction data based on the NUPEC Boiling Water Reactor Full-size Fine-mesh Bundle Test (BFBT) benchmark. Here, we found that augmenting the training dataset using VAEs has improved the DNN model’s predictive accuracy, improved the prediction confidence intervals, and reduced the prediction uncertainties.
Bayesian neural network↗
A Markov framework for generalized post-event systems recovery modeling: From single to multihazards
State-dependent models can be used to represent the system recovery process as a series of stochastic transitions from lower to higher functional states. However, the applications of these models have been limited in scope and there is a lack of a generalized recovery modeling framework. A generalized framework would permit a robust forecasting of systems and system-of-systems recovery under multiple hazards, and more broadly, would contribute to community disaster preparedness. This paper develops a generalized post hazard-event recovery modeling framework based on state-dependent Markov-type processes. We then apply the proposed framework to solve a spectrum of problems that range from hind-casting single-system recovery following a single hazard event to forecasting post-event trajectories under multiple hazards and modeling the recovery of a system-of-systems. First, Markov chains are used to hind-cast the observed recovery for a portfolio of buildings affected by the 2014 South Napa, California, earthquake. Next, Markov processes are used to formulate a parametric post hazard-event recovery model, which can be updated using Bayesian statistics when relevant datasets become available. Semi-Markov processes are then used to develop a more general model of single hazard recovery, which accounts for the intensity of the loading and level of damage caused by the event. Semi-Markov processes with non-renewal features are then used to account for multihazard interactions in a post-event recovery model, and applied to a case study that involves a community in Charleston, South Carolina. Lastly, Markov-type processes are combined with Bayesian networks to model the recovery of residential, commercial, educational, and industrial buildings (system-of-systems) following a hazard event. Overall, these applications demonstrate the versatility of the Markov framework towards handling recovery problems with varying levels of complexity.
42 ENGINEERING↗
Recovering the effects of subgrid heterogeneity in simulations of radionuclide transport through fractured media
Groundwater flow and contaminant transport through fractured media can be simulated using Discrete Fracture Network (DFN) models which provide a natural description of structural heterogeneity. However, this approach is computationally expensive, with the large number of intersecting fractures necessitated by many real-world applications requiring modeling simplifications to be made for calculations to be tractable. Upscaling methods commonly used for this purpose can result in some loss of local-scale variability in the groundwater flow velocity field, resulting in underestimation of particle travel times, transport resistance and retention in transport calculations. In this paper, a transport downscaling algorithm to recover the transport effects of heterogeneity is tested on a synthetic Brittle Fault Zone model, motivated by the problem of large safety assessment calculations for geological repositories of spent nuclear fuel. We show that the variability in the local-scale velocity field which is lost by upscaling can be recovered by sampling from a library of DFN transport paths, accurately reproducing DFN transport statistic distributions and radionuclide breakthrough curves in an upscaled model.
54 ENVIRONMENTAL SCIENCES↗
Data-driven estimation of energy consumption for electric bus under real-world driving conditions
Reliable and accurate estimation of an electric bus’s instantaneous energy consumption is critical in evaluating energy impacts of planning and control of electric bus operations. In this study, we developed machine learning-based long short-term memory (LSTM) and artificial neural network (ANN) models to estimate 1 Hz energy consumption of electric buses based on continuous monitoring data of electric buses in Chattanooga, Tennessee, in 2019 and 2020. We propose a data-partitioning algorithm to separate energy charging and discharging modes before applying data-driven estimation models. Here, a K-fold cross-validation-based model selection process was conducted to identify the optimal model structure and input variables in terms of prediction accuracy. The estimation results show the predicted mean absolute percentage error rates of LSTM and ANN models were 3% and 5%, respectively. We compared the proposed models with existing models in the literature based on the same testing data to demonstrate the predictability of our models.
Artificial neural network↗
Enabling fast charging of lithium-ion batteries through secondary- /dual- pore network: Part I - Analytical diffusion model
Battery performance is strongly correlated with electrode microstructural properties. Enabling fast charging of lithium-ion batteries requires improved through-plane ionic diffusion that can be achieved through, among other strategies, structured electrodes with a secondary- or dual-pore network (SPN). In this work, an analytical model investigates the impact of such an SPN on ionic diffusion with a composite electrode, considering various pore-channel geometries and comparing to standard electrodes with identical gravimetric- and volumetric-specific theoretical capacities. Relevant SPN design parameters and tortuosity coefficients are identified according to three optimization objectives that aim to balance the improved overall through-plane diffusion, thanks to the coarse aligned channels, and degraded in-plane diffusion because of the porous matrix densification required to maintain gravimetric- and volumetric-specific theoretical capacities. The model indicates that a relatively low amount of SPN is required and that electrodes with high through-plane tortuosity and low in-plane tortuosity benefit most from such architecture.
25 ENERGY STORAGE↗
An Engineered Glutamate in Biosynthetic Models of Heme-Copper Oxidases Drives Complete Product Selectivity by Tuning the Hydrogen-Bonding Network
Explore the source record for details and available documents.
99 GENERAL AND MISCELLANEOUS↗
Enhancing Lattice Kinetic Schemes for Fluid Dynamics with Lattice-Equivariant Neural Networks
A new class of equivariant neural networks is presented, hereby dubbed lattice-equivariant neural networks (LENNs), designed to satisfy local symmetries of a lattice structure. The approach develops within a recently introduced framework aimed at learning neural network-based surrogate models’ lattice Boltzmann collision operators. Whenever neural networks are employed to model physical systems, respecting symmetries and equivariance properties has been shown to be key for accuracy, numerical stability, and performance. Here, hinging on ideas from group representation theory, trainable layers are defined whose algebraic structure is equivariant with respect to the symmetries of the lattice cell. In this work, the presented method naturally allows for efficient implementations, in terms of both memory usage and computational costs, supporting scalable training/testing for lattices in two spatial dimensions and higher (in which the size of symmetry group grows). The approach is validated and tested considering 2D and 3D flowing dynamics, both in laminar and turbulent regimes. It is compared with group-averaged-based symmetric networks and with plain, nonsymmetric, networks, showing how the presented approach unlocks the (a posteriori) accuracy and training stability of the former models and the train/inference speed of the latter networks. (LENNs are about one order of magnitude faster than group-averaged networks in 3D.) The work in this paper opens toward practical use of machine learning-augmented lattice Boltzmann CFD in real-world simulations.
97 MATHEMATICS AND COMPUTING↗
Analysis of predicted mean vote-based model predictive control for residential HVAC systems
Model Predictive Control (MPC) is an advanced process control method that has attracted much attention in building heating, ventilation, and air conditioning (HVAC) systems. Here, this paper analyzes the optimal precooling performance in residential buildings using MPC with two different comfort indices, namely, temperature and predicted mean vote (PMV). It first formulates, for each comfort index, an optimization problem that accounts for different factors, such as weather, home thermal condition, prediction horizon, time-of use (TOU) utility rate, and rated cooling capacity. The problem is then solved, resulting in an MPC strategy that determines the HVAC on/off control signal and minimizes energy cost over a receding time horizon while maintaining thermal comfort. The energy performance difference between temperature-based and PMV-based MPC strategies is subsequently investigated, especially in light of the interior wall surface temperature and under different combinations of the factors. Extensive simulation results demonstrated that the proposed MPC strategies are adaptive and their performances depend primarily on weather, home thermal condition, and prediction horizon, while the impact of TOU utility rate and rated cooling capacity is relatively small. Because the PMV-based MPC strategy can take advantage of the lower interior wall surface temperature due to precooling, it resulted in 8–45% cost savings for the scenarios investigated and an average increase of 0.042–0.113 in the absolute value of the PMV index compared to the temperature-based MPC strategy.
32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
Spectroscopic Signatures of MQ-Resins in Silicone Elastomers
Polysiloxane elastomers have a large application space due to their versatile cross-linking chemistry and highly tunable physical and mechanical properties. One approach for improving the mechanical integrity of commercial polysiloxane “silicone” elastomers while maintaining their optical transparency is the addition of small, silicone-resin molecules to the network. However, both the poly(dimethylsiloxane) (PDMS) network and the silicone-resin particles have an amorphous structure and complex chemistry, which makes the characterization of their structural properties and segmental network dynamics difficult. Here, we report the synthesis and characterization of a series of model silicone networks modified with a specific class of silicone-resin known as MQ-resin using Raman and advanced nuclear magnetic resonance (NMR) spectroscopy methods. Raman spectroscopy was successfully used to quantify the contribution of the MQ-resin to the network, to determine the type of MQ-resin present in the network, and to investigate the completeness of the network cross-linking reaction. Furthermore, solid-state and 1H double-quantum (DQ) NMR spectroscopies were used not only as a detection method for the MQ-resins but also to quantify changes in the segmental dynamics of the network as a function of MQ-resin concentration. The combination of Raman and NMR spectroscopies describes a series of samples where the MQ-resin particles and PDMS chains maintain their independent segmental dynamics up to high concentrations of MQ-resins (40–50% MQ), where the physical properties of the resin dominate the physical properties of the overall network. The results from our spectroscopic analyses are consistent with the results from macroscopic characterization techniques such as solvent uptake and mechanical testing. The spectroscopic insights into the structure–property relationships of PDMS-MQ composites presented in this study are a valuable tool not only for the synthesis and reverse engineering of future generations of commercial silicone elastomers but also for understanding the mechanisms of aging and degradation over the material lifetime.
36 MATERIALS SCIENCE↗
A Deep Learning Filter for the Intraseasonal Variability of the Tropics
Abstract This paper presents a novel application of convolutional neural network (CNN) models for filtering the intraseasonal variability of the tropical atmosphere. In this deep learning filter, two convolutional layers are applied sequentially in a supervised machine learning framework to extract the intraseasonal signal from the total daily anomalies. The CNN-based filter can be tailored for each field similarly to fast Fourier transform filtering methods. When applied to two different fields (zonal wind stress and outgoing longwave radiation), the index of agreement between the filtered signal obtained using the CNN-based filter and a conventional weight-based filter is between 95% and 99%. The advantage of the CNN-based filter over the conventional filters is its applicability to time series with the length comparable to the period of the signal being extracted. Significance Statement This study proposes a new method for discovering hidden connections in data representative of tropical atmosphere variability. The method makes use of an artificial intelligence (AI) algorithm that combines a mathematical operation known as convolution with a mathematical model built to reflect the behavior of the human brain known as artificial neural network. Our results show that the filtered data produced by the AI-based method are consistent with the results obtained using conventional mathematical algorithms. The advantage of the AI-based method is that it can be applied to cases for which the conventional methods have limitations, such as forecast (hindcast) data or real-time monitoring of tropical variability in the 20–100-day range.
Stan, Cristiana↗
GraphAlign: Graph-Enabled Machine Learning for Seismic Event Filtering
This report summarizes results from a 2 year effort to improve the current automated seismic event processing system by leveraging machine learning models that can operated over the inherent graph data structure of a seismic sensor network. Specifically, the GraphAlign project seeks to utilize prior information on which stations are more likely to detect signals originating from particular geographic regions to inform event filtering. To date, the GraphAlign team has developed a Graphical Neural Network (GNN) model to filter out false events generated by the Global Associator (GA) algorithm. The algorithm operates directly on waveform data that has been associated to an event by building a variable sized graph of station waveforms nodes with edge relations to an event location node. This builds off of previous work where random forest models were used to do the same task using hand crafted features. The GNN model performance was analyzed using an 8 week IMS/IDC dataset, and it was demonstrated that the GNN outperforms the random forest baseline. We provide additional error analysis of which events the GNN model performs well and poorly against concluded by future directions for improvements.
58 GEOSCIENCES↗
Comparative Analysis of HEATNETS for Geothermal Network Performance: Preprint
Thermal energy networks (TENs), also known as 5th generation district energy systems, or more specifically geothermal networks when exchanging heat with geothermal boreholes, are an important technology for decarbonization. In these networks an ambient loop connects buildings and thermal sources, such as a borehole field, to exchange energy and maintain a desired loop temperature. Water-source heat pumps are used at the buildings to connect to the ambient or thermal loop to meet to the building heating and cooling loads and maintain comfort. A semi-transient, reduced-order technical model and techno-economic model, called HEATNETS, has been developed at NREL that captures the flow of energy around a TEN. In this work, a comparison of the HEATNETS technical model and a well-known coding platform used for modeling geothermal networks, TRNSYS, has been completed for a proposed geothermal network as a verification process. Hourly data provided from the TRNSYS simulation included building loads, pumping power, heat pump power, temperature entering and leaving the borehole field, and mass flow rates. The hourly borehole temperatures were used to create a linear regression model utilized in HEATNETS to estimate the borehole field heat exchange. The building loads and mass flow rates were direct inputs to HEATNETS while the pumping power, heat pump power, borehole temperatures, and coefficients of performance were all simulated and calculated by HEATNETS, allowing for direct comparison of the thermal energy transfer, rather than also comparing control systems responses. HEATNETS considers the full process from design inputs to economic outputs and can provide modeling options for high-level initial system design and operational optimization. This study shows that HEATNETS, while not intended to replace other modeling tools, can be a unique modeling tool for the performance of a full geothermal network system.
15 GEOTHERMAL ENERGY↗
Modeling MTS pyrolysis and SiC deposition kinetics using principal component analysis and neural networks
Accurate chemical kinetics modeling is crucial for improving the efficiency of chemical processing and synthesis of ceramic matrix composites. Detailed kinetic models are computationally expensive due to the large number of transported chemical species, while the simplified physics-based models, such as single-step global mechanisms, are efficient but often overlook key chemical intermediates and pathways. Recent deep learning approaches promise accurate and cost-effective models. Yet, they require additional closures for the transported nonlinear latent variables, complicating integration with existing solvers. In this work, we develop a hybrid linear—nonlinear reduced model for silicon carbide deposition from methyltrichlorosilane precursor by combining principal component analysis (PCA) and autoencoder (AE) neural network (NN) approaches. PCA is used to identify a smaller set of linear transport variables, enabling direct reuse of conventional transport solvers. NNs then reconstruct the full chemical state from these reduced variables. We demonstrate the method on a chemical vapor deposition reactor—comprising a gas-phase pyrolysis plug flow reactor and a heterogeneous surface reactor—over a wide range of temperatures, pressures, and residence times. Our PCA–AE model achieves high accuracy with only five transported scalars, achieving an eightfold cost reduction compared to detailed mechanisms, in both a priori (using data from the test set only) and a posteriori (coupled with a differential equation solver). In conclusion, notable errors arise primarily near training domain boundaries and for long residence times, indicating the need for domain shift indicators and better long-horizon predictions in future reduced chemistry model development.
autoencoder neural networks↗
Neural network denoising of HED x-ray images, with an introduction to neural networks [Slides]
We created and trained a new neural network denoiser model using estimates of noise in the data. We showed that our denoiser significantly reduces the combined blur and noise on the training and testing set. We demonstrated the efficacy of neural network denoisers for reducing noise on HED x-ray images. Changes to network architecture and improvements to the noise model in training could improve the model.
47 OTHER INSTRUMENTATION↗