SEARCH · Engineering Papers
Results for “Mathematics and Computing, Computer science”
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.
Backpropagation-based learning with local derivative approximation and memory replay in biologically plausible neural systems
When learning, the brain modifies individual synaptic connections to reach a desired behavior. Animal and human brains have been shown to be incredibly capable of learning complex and varied functions across a wide variety of tasks. In recent years, artificial neural networks, inspired by human and animal brains, have shown great capabilities in learning a wide variety of difficult tasks. However, artificial neural networks primarily teach themselves through the use of backpropagation, a learning method which has no clear analogue within the brain. Additionally, Artificial Neural Networks primarily use continuous activation functions, which differ significantly from the spiking neuronal behavior present in the brain. In this paper, we discuss and demonstrate a biologically plausible learning method that approximates backpropagation through two techniques on Spiking Neural Networks. First, we show that the local temporal derivatives that are necessary for backpropagation can be approximately recovered through reconstruction using spike timings. Second, we show that through learning during a sleep phase, inspired by neuroscience research into memory replay, the localized parallel feedback path can learn to approximate the derivative through the forward path weight matrix, thus solving the weight transport problem. Lastly, we demonstrate that the combination of these two methods can approach or exceed the accuracy of backpropagation-based methods for a variety of neuromorphic vision tasks while maintaining biological plausibility.
Multiphysics simulation of TRISO fuel compacts and the effects of homogenization on silver release predictions
This work studies the impact of explicit and homogenized modeling approaches on the multiphysics simulation of TRistructural ISOtropic (TRISO) fuel compacts in prismatic High Temperature Gas Reactors (HTGRs) and the silver release predictions. TRISO fuel compacts exhibit complex double heterogeneity that significantly affects heat conduction, neutron transport, and fission products release. In this work, we use Cardinal, a multiphysics tool based on the Multiphysics Object-Oriented Simulation Environment (MOOSE) framework, to couple neutron transport and heat conduction. OpenMC is used for neutron transport and the MOOSE heat transfer module is used for the heat conduction. Then, we use BISON for the silver release predictions. Two different modeling approaches—explicit modeling of individual TRISO particles and homogenized representation using effective thermal properties—are compared at high TRISO packing fractions (20% and 40%) across varying power densities. Results demonstrate that homogenization significantly under-predicts peak temperatures, for the case of high power/TRISO, there is a difference of 83.73 K in the maximum temperature. Additionally, homogenization underestimates the silver release predictions compared to explicit modeling. Specifically, the temperature and power density differences lead to significant differences in silver release predictions. This work demonstrates the importance of accurately modeling heterogeneity of the TRISO particles to reliably predict fission product release and assess reactor safety margins.
Rotor blade imbalance fault detection for variable-speed marine current turbines via generator power signal analysis
Marine hydrokinetic (MHK) turbines extract renewable energy from oceanic environments. However, due to the harsh conditions that these turbines operate in, system performance naturally degrades over time. Thus, ensuring efficient condition-based maintenance is imperative towards guaranteeing reliable operation and reduced costs for marine hydrokinetic power. This work proposes a novel framework aimed at identifying and classifying the severity of rotor blade pitch imbalance faults experienced by marine current turbines (MCTs). In the framework, a Continuous Morlet Wavelet Transform (CMWT) is first utilized to acquire the wavelet coefficients encompassed within the 1P frequency range of the turbine's rotor shaft. From these coefficients, several statistical indices are tabulated into a six-dimensional feature space. Next, Principle Component Analysis (PCA) is employed on the resulting feature space for dimensionality reduction, and then the application of a K-Nearest Neighbor (KNN) machine learning algorithm is utilized for fault detection and severity classification. The effectiveness of the proposed framework is validated using a high-fidelity MCT numerical simulation platform, where results demonstrate that the presence of a pitch imbalance fault can be accurately detected 100% of the time and correctly classified based upon severity more than 97% of the time.
The role of geographical spreaders in infectious pattern formation and front propagation speeds
The pattern formation and spatial spread of infectious populations are investigated using a kernel-based Susceptible–Infectious–Recovered (SIR) model applicable across a wide range of basic reproduction numbers R O . The goal is to examine the role of geographical spreaders on transient spatial pattern formation of infectious populations and the associated maximum invasive front speeds c max . In the simulations conducted here, geographical spreaders are defined as a portion of the infected population Φ experiencing high mobility between identical communities. The spatial organization of the infected population and c max are determined when the infections are randomly initiated in space within multiple communities. For small but finite , scaling analysis and numerical simulations in 1-dimension suggest that when the spreading kernel is Gaussian-shaped, where is the inverse of the infectious duration. This finding for agrees with a diffusion-based representation of mobility in 1-D. Numerical simulations in 2-D across wide-ranging suggest that , the variance of the spatial kernel describing mobility of long-distance geographical spreaders across communities, determines the spatial organization of infections across communities. When (long-distance mobility, where is the minimum spatial extent defining adjacent communities), the infectious population will experience a transient but spatially coherent pattern with a wavelength that can be derived from the spreading kernel properties. Moreover, the 2-D simulations for the bounded kernel suggest that attainment of is also dictated by but the magnitude is not sensitive to unlike diffusion-based models.
Promoting regulatory acceptance of combined ion and neutron irradiation testing of nuclear reactor materials: Modeling and software considerations
As the needs for the nuclear energy industry continue to evolve in the 21st century, timely adoption of new technological solutions acceptable to regulatory agencies is critical. Quantitative prediction of radiation damage in materials and its impact on mechanical properties is a key component of licensing and regulatory decisions regarding nuclear power plants. Accelerated testing methodologies such as combined ion and neutron irradiation data sets are crucial for the development and deployment of new materials and new manufacturing methods (e.g., additive manufacturing). However, regulatory acceptance of accelerated testing methodologies is necessary for their adoption. Further, the present work discusses the fundamental basis for comparing ion- and neutron-induced material microstructures, the theory behind interpreting radiation damage across length and time scales and radiation types, and the codes, standards, and quality assurance concerns surrounding different modeling methods and software. In particular, recommendations are given as to the path forward that will enable national laboratories, academia, and industry to develop the modeling and software basis for regulatory acceptance of the combined use of ion and neutron irradiation for material performance evaluation.
Global Electrification of Vehicles and Intertwined Material Supply Chains of Cobalt, Copper and Nickel
Electric vehicles (EVs) will be an important part of a low-carbon economy. Future EV adoption depends on overcoming multiple barriers, including charging infrastructure, sufficient electricity generation, and the availability of battery raw materials at acceptable costs, including cobalt, copper and nickel – which typically are co-produced. Various EV projections have been made, along with associated requirements for battery raw materials (demand). There have been separate assessments of the availability of raw materials (supply). However, integrated assessments of supply and demand and their interaction over time are lacking to support decision-making. Using a dynamic market model that reflects the co-produced nature of cobalt, copper and nickel production as well as recycling, we estimate global output of these three metals under different EV growth scenarios from 2020 to 2040. In this work, results suggest that starting in 2025, replacement demand for battery raw materials becomes important. Thus, estimates for future material demand that consider only demands from new EVs are incomplete and misleading. EV deployment and associated increased demand for raw materials will have a greater impact on prices for cobalt than nickel and, in turn, for nickel than copper. High EV deployment scenarios will be difficult to achieve without larger supplies of copper, nickel and cobalt than forthcoming under the conditions modeled here, which anticipates a six-year time lag between a demand increase and the expansion of metal-production capacity. Shortening battery lifetimes and improving collection and recycling rates increase secondary supply, but the extent to which these factors support additional EV deployment is mixed.
An integrated supply chain analysis for cobalt and rare earth elements under global electrification and constrained resources
As global electrification becomes more prominent, demand is projected to increase for electric vehicles (EVs), their components (EV batteries, rare earth traction motors), and their raw materials (cobalt, rare earths). To analyze the supply chain constraints associated with these components and raw materials, a system dynamics model was developed to simulate the interactions between the rare earth and cobalt supply chains. The results showed that under current understanding of raw material resources, production of the projected volume of EVs could be inadequate. To meet the aggressive EV volume sale projections set by the International Energy Agency, EV manufacturers will need to implement a large portion of traction motors without rare earths and utilize non-cobalt batteries as much as possible. On the supply side, a much faster ramp-up than what being modeled would be needed to meet much higher demand than current market forecasts.
Anisotropic grain boundary area and energy distributions in tungsten
Describing microstructure evolution in tungsten requires a quantitative description of the anisotropic grain boundary energy. Here we present a grain boundary energy function for tungsten that specifies the energy of an arbitrary boundary given its five macroscopic crystallographic parameters. A comparison of measured grain boundary areas and the grain boundary energies given by the function at the Σ11, Σ17b, and Σ33a misorientations, which are problematic to determine by measurement or atomistic calculations, reveals inverse correlations that are similar to what have been observed in other metals.
Parameter-free prediction of phase transition in PbTiO 3 through combination of quantum mechanics and statistical mechanics
Thermodynamics of ferroelectric materials and their ferroelectric to paraelectric (FE-PE) transitions is commonly described by the phenomenological Landau theory and more recently by effective Hamiltonian and various potentials, all with model parameters fitted to experimental or theoretical data. Here, in this work, we show that the zentropy theory, which considers the total entropy of a system as a weighted sum of entropies of configurations that the system may experience and the statistical entropy among the configurations, can predict the FE-PE transition without fitting parameters. For PbTiO 3 , the configurations are identified as the FE configurations with 90- or 180° domain walls in addition to the ground state FE configuration without domain wall. With the domain wall energies predicted from first-principles based on density functional theory in the literature as the only inputs, the FE-PE transition for PbTiO 3 is predicted showing remarkable agreement with experiments, unveiling the microscopic fundamentals of the transition.
ADIOS 2: The Adaptable Input Output System. A framework for high-performance data management
Not Available
A primer on artificial intelligence in plant digital phenomics: embarking on the data to insights journey
Artificial intelligence (AI) has emerged as a fundamental component of global agricultural research that is poised to impact on many aspects of plant science. In digital phenomics, AI is capable of learning intricate structure and patterns in large datasets. We provide a perspective and primer on AI applications to phenome research. We propose a novel human-centric explainable AI (X-AI) system architecture consisting of data architecture, technology infrastructure, and AI architecture design. We clarify the difference between post hoc models and 'interpretable by design' models. We include guidance for effectively using an interpretable by design model in phenomic analysis. We also provide directions to sources of tools and resources for making data analytics increasingly accessible. In conclusion, this primer is accompanied by an interactive online tutorial.
A Denoising Autoencoder for Improved Kikuchi Pattern Quality and Indexing in Electron Backscatter Diffraction
The rapid collection and indexing of electron diffraction patterns as produced via electron backscatter diffraction (EBSD) has enabled crystallographic orientation and structural determination, as well as additional property-determining strain and dislocation density information with increasing speed, resolution, and efficiency. Pattern indexing quality is reliant on the noise of the collected electron diffraction patterns, which is often convoluted by sample preparation and data collection parameters. EBSD acquisition is sensitive to many factors and thus can result in low confidence index (CI), poor image quality (IQ), and improper minimization of fit, which can result in noisy datasets and misrepresent the microstructure. In an attempt to enable both higher speed EBSD data collection and enable greater orientation fit accuracy with noisy datasets, an image denoising autoencoder was implemented to improve pattern quality. Here, we show that EBSD data processed through the autoencoder results in a higher CI, IQ, and a more accurate degree of fit. In addition, using denoised datasets in HR-EBSD cross correlative strain analysis can result in reduced phantom strain from erroneous calculations due to the increased indexing accuracy and improved correspondence between collected and simulated patterns.
Modeling and Optimization of a Rotating Packed Bed Contactor with a Tetraamine-Appended Metal−Organic Framework for CO 2 Capture
A potential contactor technology for sorbent-based CO 2 capture is the rotating packed bed that contains separate sections for continuous adsorption and desorption. A heat exchanger can be embedded to remove heat in the adsorption section and add heat in the desorption section. In this work, we develop a two-dimensional (2D) model of a rotating packed bed for use in CO 2 capture applications. Mass and energy balances for the model are developed based on a Ljungström-type air preheater, which accounts for the counter-current axial flow of gas phases in separate sections of the bed and the rotation of a solid sorbent, which cycles between adsorption and desorption sections. The sorbent used for this analysis is the tetraamine-appended metal−organic framework Mg 2 (dobpdc)(3−4− 3), chosen for its stability and affinity for CO 2 at low partial pressures, such as those from a natural gas power plant source. An optimization problem is solved that considers the trade-off between maximizing the productivity of the bed and minimizing energy consumption. Maximum productivity and minimum energy are found to be 8.53 kg/h/m 3 and 3.84 MJ/kg, respectively, when these objectives are optimized independently. It is observed that the flue gas pressure and bed rotational speed are the desired operating variables to vary for model-based design of experiments to reduce uncertainty in parameter estimation, as these two variables yielded the most information content based on the Fisher information matrix.
Surfactant-Specific AI-Driven Molecular Design: Integrating Generative Models, Predictive Modeling, and Reinforcement Learning for Tailored Surfactant Synthesis
Molecular design is a critical aspect of various scientific and industrial fields, where the properties of molecules hold significant importance. In this study, a 3-fold methodology design is presented that leverages the power of generative artificial intelligence (AI), predictive modeling, and reinforcement learning to create tailored molecules with desired properties. This model synergistically combines deep learning techniques with Self-Referencing Embedded Strings (SELFIES) molecular representation to build a generative model that generates valid molecules and a graphical neural network model that accurately forecasts molecular properties. The Variational Autoencoder (VAE) coupled with reinforcement learning helps refine molecule generation based on targeted attributes. Data from an experimental study involving surfactants were used to test the framework. A validation of the structural integrity of the molecules generated was conducted, and Tanimoto similarities were used to quantify the similarity and diversity between the original and generated molecular structures. Also, saliency maps for the generated surfactants were produced to identify the features explaining the property values. Lastly, molecular dynamics simulations were used to validate the stability of the generated molecules. The results showed that the proposed framework can effectively produce valid molecules within the set property threshold value.
Economic NMPC for a Reversible Solid Oxide Cell
Reversible solid oxide fuel cells (rSOCs) offer the flexibility to operate in tandem with the electric grid by switching between fuel cell and electrolysis modes based on real-time electricity prices. However, their complex, tightly coupled dynamic behavior poses significant challenges in determining optimal operating strategies. In this work, we present an economic nonlinear model predictive control (E-NMPC) framework to optimize the operation of rSOCs. The proposed E-NMPC is applied to a detailed rSOC flowsheet model that includes a utility scale rSOC module as well as balance-of-plant equipment necessary for thermal management. Our results demonstrate that in fuel cell mode, the E-NMPC strategy reduces hydrogen consumption compared to conventional set-point tracking NMPC, while maintaining the same level of electricity output. Also, in electrolysis mode, the E-NMPC yields a marginal improvement in hydrogen production. In addition, we explore the integration of a battery with the rSOC system to enhance flexibility in meeting electricity production and consumption targets.
Protein Surface Printer for Exploring Protein Domains
The surface of proteins is vital in determining protein functions. Herein, a program, Protein Surface Printer (PSP), is built that performs multiple functions in quantifying protein surface domains. Two proteins, PETase and cytochrome P450, are used to validate that the program supports atomistic simulations with different combinations of programs and force fields. A case study is conducted on the structural analysis of the spike proteins of SARS-CoV-2 and SARS-CoV and the human cell receptor ACE2. Although the surface domains of both spike proteins are highly similar, their receptor-binding domains (RBDs) and the O-linked glycan domains are structurally different. The O-linked glycan domain of SARS-CoV-2 is highly positively charged, which may promote binding to negatively charged human cells.
High-Throughput Virtual Screening and Validation of a SARS-CoV-2 Main Protease Noncovalent Inhibitor
Not Available