Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Reduce order modeling”

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 1,351 records · Page 75
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Assessing the Feasibility of Bordered Block Diagonal Reordering in Power System Matrices using Fully Convolutional Network

In electromagnetic transient (EMT) simulations for power systems and inverter-based resources (IBRs), the arrangement of states within the system's linear equations, represented by matrix A in Ax=b, is critical. The state ordering in matrix A can highlight distinct characteristics of the system's graph, and identifying an optimal state ordering is crucial for efficient computation. The choice of state ordering, however, is dependent on the solver used, as each solver may perform optimally with different matrix patterns. With a wide array of matrix reordering algorithms available, selecting the most suitable one becomes challenging without insights into the matrix's ideal configuration. To address this, the paper proposes a fully convolutional network (FCN) to evaluate the reordering potential of the A matrix into a bordered block diagonal (BBD) pattern, which is commonly observed in power system and IBR modeling. The FCN's assessment aims to streamline the solver's operation, which in turn could substantially reduce the computational time required to find a solution.

Xia, Qianxue↗
DOE OSTI2020

Probing the Effects of NO x and SO x Impurities on Oxy-Fuel Combustion in Supercritical CO 2 : Shock Tube Experiments and Chemical Kinetic Modeling

The direct-fired supercritical carbon dioxide cycles are one of the most promising power generation methods in terms of their efficiency and environmental friendliness. Two important challenges in implementing these cycles are the high pressure (300 bar) and high CO 2 dilution (>80%) in the combustor. The design and development of supercritical oxy-combustors for natural gas require accurate reaction kinetic models to predict the combustion outcomes. The presence of a small amount of impurities in natural gas and other feed streams to oxy-combustors makes these predictions even more complex. During oxy-combustion, trace amounts of nitrogen present in the oxidizer is converted to NO x and gets into the combustion chamber along with the recirculated CO 2 . Similarly, natural gas can contain a trace amount of ammonia and sulfurous impurities that get converted to NO x and SO x and get back into the combustion chamber with recirculated CO 2 . In this work, a reaction model is developed for predicting the effect of impurities such as NO x and SO x on supercritical methane combustion. The base mechanism used in this work is GRI Mech 3.0. H 2 S combustion chemistry is obtained from Bongartz et al. while NOx chemistry is from Konnov. The reaction model is then optimized for a pressure range of 30–300 bar using high-pressure shock tube data from the literature. It is then validated with data obtained from the literature for methane combustion, H 2 S oxidation, and NO x effects on ignition delay. The effect of impurities on CH 4 combustion up to 16 atm is validated using NO x -doped methane studies obtained from the literature. In order to validate the model for high-pressure conditions, experiments are conducted at the UCF shock tube facility using natural gas identical mixtures with N 2 O as an impurity at ~100 bar. Current results show that there is a significant change in ignition delay with the presence of impurities. A comparison is made with experimental data using the developed model and predictions are found to be in good agreement. The model developed was used to study the effect of impurities on CO formation from sCO 2 combustors. It was found that NO x helps in reducing CO formation while the presence of H 2 S results in the formation of more CO. The reaction mechanism developed herein can also be used as a base mechanism to develop reduced mechanisms for use in CFD simulations.

Energy & Fuels↗
DOE OSTI2022

Self‐Assembled LuFeO 3 /LuFe 2 O 4 Heterostructure with Emergent Ferroic Orderings

Abstract Achieving self‐assembling structures that exhibit emergent properties and functionalities is one of the holy grails in nanotechnology and holds great promise for a wide range of potential applications. Herein, a self‐assembled LuFeO 3 –(1/3)LuFe 2 O 4 –LuFeO 3 heterostructure that consists of two model multiferroics, i.e., hexagonal LuFeO 3 and rhombohedral LuFe 2 O 4 , is reported in single‐layer epitaxial LuFeO 3 thin films. Atomic‐scale electron microscopy investigations reveal the spontaneous organization of ferroelectric polarization in each building block, which forms a series of polar configurations. Specifically, the charge and lattice of neighboring LuFeO 3 and LuFe 2 O 4 blocks are closely coupled with competing degrees of freedom. A combination of the quantitative analysis of atomically resolved images and spectroscopy with theoretical calculations determines how charge‐ordering‐induced polarization accommodates charge naturality across the interfaces to reduce the system energy and unravels the formation mechanism of this heterostructure. The results point to the possibilities of generation and control of multiferroic self‐assembled systems, which may provide a basis for engineering novel materials with emergent properties.

Deng, Shiqing↗
DOE OSTI2021

Computational Evaluation of Thermal Response of Open-Cell Foam With Circular Pore (Computional Evaluation of Thermal Response of Open-Cell Foam with Circular Pore)

The evaluation of effective material properties in heterogeneous materials (e.g., composites or multicomponent structures) critically relevant to a wide spectrum of applications, including nuclear power, electronic packaging, flame retardants, hypersonics, and gas turbine power. The work described in this paper is centered around the numerical assessment of the thermal behavior of porous materials obtained from finite element thermal modeling and simulation. Here, two-dimensional, steady state analyses were performed on unit cells with centered, circular pores using a second order accurate Galerkin finite element method (FEM). The effective thermal conductivities of the porous systems were examined, encompassing a range of porosities from 4.9% to 60.1%. The geometries of the models were generated based on ordered circular pores for each modeled porosity level. The system response quantity (SRQ) under investigation was the dimensionless effective thermal conductivity across the unit cell. The dimensionless effective thermal conductivity was compared across all simulated cases, producing a trend between porosity and effective thermal conductivity. In the presented investigation, the method of manufactured solutions (MMS) was used to perform code verification, and the grid convergence index (GCI) was employed to estimate discretization uncertainty as solution verification. Code verification concluded an approximately second order accurate Galerkin FEM solver. It was found that the introduction of porosity to the unit cell material structure reduces effective thermal conductivity, as anticipated. Numerical results obtained in this study are compared to an analytical solution and to a sample of empirical data.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Three-dimensional core-collapse supernova models with phenomenological treatment of neutrino flavor conversions

Abstract We perform three-dimensional supernova simulations with a phenomenological treatment of neutrino flavor conversions. We show that the explosion energy can increase to as high as $\sim 10^{51}$ erg depending on the critical density for the onset of flavor conversions, due to a significant enhancement of the mean energy of electron antineutrinos. Our results confirm previous studies showing such energetic explosions, but for the first time in three-dimensional configurations. In addition, we predict neutrino and gravitational wave (GW) signals from a nearby supernova explosion aided by flavor conversions. We find that the neutrino event number decreases because of the reduced flux of heavy-lepton neutrinos. In order to detect GWs, next-generation GW telescopes such as Cosmic Explorer and the Einstein Telescope are needed even if the supernova event is located at the Galactic Center. These findings show that the neutrino flavor conversions can significantly change supernova dynamics and highlight the importance of further studies on the quantum kinetic equations to determine the conditions of the conversions and their asymptotic states.

Mori, Kanji↗
DOE OSTI2019

Highly Efficient Separation/Recycling Palladium(II) Ions from Aqueous Solutions by Silica Gel-Coated Graphene Oxide Modified with Mercapto Groups

The separation and recovery of precious metals from secondary resources is an extremely important and challenging task. Herein, a novel sorbent is successfully prepared by chemical grafting and utilized for the extraction and separation of Pd(II) ions from aqueous solutions. Various characterization techniques, including Fourier transform infrared spectroscopy, scanning electron microscopy, Brunauer–Emmett–Teller, X-ray photoelectron spectroscopy, and element analysis, are employed to study the structure, morphology, porous nature, and chemical composition of the as-prepared SiO{sub 2}@GO-SH nanocomposites. The adsorption of Pd(II) ions on the SiO{sub 2}@GO-SH can be described by the pseudo-second-order model and Langmuir isotherm model. The adsorption equilibrium has been attained within 90 min and the maximum adsorption capacity of 423.2 mg g{sup −1} was attained at pH 3.5 and T = 308 K. The results reveal that SiO{sub 2}@GO-SH exhibits an excellent adsorption performance towards Pd(II) ions. In addition, we have proposed the adsorption mechanism for Pd(II) ions on the SiO{sub 2}@GO-SH surface. The adsorption of Pd(II) in SiO{sub 2}@GO-SH is a chemisorption process, where partial adsorbed Pd(II) ions are reduced to Pd(0) by functional groups (-SH) in the SiO{sub 2}@GO-SH. The results indicate that SiO{sub 2}@GO-SH can serve as a promising sorbent for the efficient separation and recovery of palladium from the palladium-containing secondary resources.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Ab initio study of short-range ordering in vanadium-based disordered rocksalt structures

Disordered rocksalt Li-excess (DRX) compounds are attractive new cathode materials for Li-ion batteries as they contain resource-abundant metals and do not require the use of cobalt or nickel. Understanding the delithiation process and cation short-range ordering (SRO) in DRX compounds is essential to improve these promising cathode materials. Herein, we use first-principles calculations along with the cluster-expansion approach to model the disorder in DRX Li 2-x VO 3 , 0 < x < 1. We discuss the SRO of Li in tetrahedral and octahedral sites, and the order in which Li delithiates and V oxidizes with respect to local environments. We reveal that the number of nearest-neighbor V dictates the order of delithiation from octahedral sites and that V is oxidized in a manner that minimizes the electrostatic interactions among V. In conclusion, our results provide valuable insight for tailoring the performance of V-based DRX cathode materials in general by controlling the SRO features that reduce energy density.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Role of microtearing mode in DIII-D and future high- β p core plasmas

This paper provides modeling evidence for a new variant of microturbulence instability that regulates the electron thermal fluxes in the transport barrier region of DIII-D "high β p " discharges. This scenario is being investigated as a possible scenario for future steady-state reactors, and thus it is essential to understand the physics processes which govern its performance in order to confidently extrapolate to future devices. Additional analysis examining the extrapolation of this scenario to ITER shows that while the new microtearing mode identified here is unlikely to be unstable in that regime (due to increased temperatures and reduced collisionality), the large shift in magnetic axis (Shafranov shift) that arises in these plasmas may still favor stabilization of common ballooning instabilities which peak on the low field side, but destabilize electrostatic slab modes on the high field side.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

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

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

machine learning↗
DOE OSTI2026

T-FSM: A Scalable Distributed Task-Based System for Frequent Subgraph Pattern Mining from a Big Graph

Finding frequent subgraph patterns in a big graph is an important problem with many applications such as classifying chemical compounds and building indexes to speed up graph queries. Since this problem is NP-hard, some recent parallel and distributed systems have been developed to accelerate the mining. However, they often have a huge memory cost, very long running time, suboptimal load balancing, poor scale-out capability, and possibly inaccurate results. In this article, we propose an efficient system called T-FSM for parallel mining of frequent subgraph patterns in a big graph. T-FSM supports a new anti-monotonic frequentness measure called Fraction-Score, which is more accurate than the widely used MNI measure. The execution engine of T-FSM supports both intra-machine parallelism and inter-machine parallelism. For intra-machine parallelism, T-FSM adopts a novel task-based execution model to ensure high multithreading concurrency, bounded memory consumption, and effective load balancing. For inter-machine parallelism, T-FSM ensures good scale-out performance with a lightweight pattern rebalancing approach that reduces workload skewness of pattern evaluations among machines. To avoid recomputing the contexts for migrated patterns, we design a novel context cache table to support concurrent and asynchronous requesting and caching of remote context data, which can timely evict and garbage collect used pattern contexts that are no longer needed to keep memory consumption bounded. Extensive experiments show that T-FSM is orders of magnitude faster than existing state-of-the-art parallel systems (more than 10×, 51×, 131×, 55× speedup over ScaleMine, DistGraph, Pangolin and Peregrine, respectively) and distributed systems (more than 42× and 88× over ScaleMine and DistGraph, respectively) for frequent subgraph pattern mining, and it scales out satisfactorily to 512 CPU cores on the Polaris supercomputer at Argonne National Laboratory.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Bending and shear improvements in 3D-printed core sandwich composites through modification of resin uptake in the skin/core interphase region

Fused deposition modeling (FDM) printed polymers are rarely used as a structural material due to anisotropic and low mechanical properties compared with conventional composites. In recent years, greater need has been expressed for recycling of materials, such as recyclable FDM, at the end of service life to reduce environmental pollution and manufacture cost. However, how the amount of resin uptake in the skin and skin/core interphase affects the bending and shear performance of the sandwich composites when replacing the low strength and ductile core (conventional core) with a high strength and brittle core (FDM printed PLA (polylactic acid) core) still remains unclear. A new manufacturing routine is needed to improve the incorporation of FDM printed polymers in composite structures. In this work, FDM printed PLA was used as core material and sandwiched between two unidirectional glass fiber reinforced polymer (GFRP) skins to form a sandwich composite by compression-molding (CM) process, which provides a good manufacturing strategy for skin/core interphase modification. The significance of the CM process is proved by investigating the effect of resin uptake on bending and in-plane/out-of-plane shear performances. Current first order shear deformation (FSDT) theory lacks a direct connection between the in-plane shear stress and out-of-shear stress in the core region of sandwich composites. With the help of DIC, a connection between the in-plane shear and the out-of-plane shear strain was built and in-plane shear properties can acquire through out-of-plane shear properties, hence reducing the redundancy of sample preparation or the need for simulation. A significant improvement was found compared with the optimized resin uptake (Optimized resin uptake range: 20.43%–22.86 wt%) 3D-printed PLA core sandwich composite and lowest performance sandwich composite (Improvement: in-plane shear strength (~34%)/modulus (~29%), out-of-plane shear strength (~25%)/modulus (~31%), specific peak bending load (~19%)). Finally, compared with balsa core sandwich composites, the 3D-printed cores are suitable for use in composite sandwich structures in many applications with a satisfactory strength-to-weight ratio.

3D-printed core sandwich composite↗
DOE OSTI2024

Stress intensity factor models using mechanics-guided decomposition and symbolic regression

The finite element method can be used to compute accurate stress intensity factors (SIFs) for cracks with complex geometries and boundary conditions. In contrast, handbook solutions act as surrogate SIF models that provide significantly faster evaluation times. However, the development of conventional surrogate SIF models relies on manual development based on low-order parameterizations. This limits surrogate model accuracy and generalizability. Here, in this paper, we develop a framework for the automated development of mechanics-guided handbook SIF solutions by using interpretable machine learning via genetic programming for symbolic regression (GPSR). Formalizing the mechanics-based approach of Raju and Newman, SIF training data is decomposed into multiple subsets. This decomposition enables parallel GPSR model development of subfunctions, each of which accounts for specific geometrical corrections with respect to a known analytical model. Using this mechanics-based approach with GPSR allows for equations to be learned with improved accuracy and reduced complexity relative to the Raju Newman equations while maintaining the inherent interpretability of mathematical expressions. In this paper, we present equations that match the complexity of the Raju Newman equations while having reduced error, as well as equations with similar errors and reduced complexity.

42 ENGINEERING↗
DOE OSTI2022

Discovering causal structure with reproducing-kernel Hilbert space ε -machines

We merge computational mechanics’ definition of causal states (predictively equivalent histories) with reproducing-kernel Hilbert space (RKHS) representation inference. The result is a widely applicable method that infers causal structure directly from observations of a system’s behaviors whether they are over discrete or continuous events or time. A structural representation—a finite- or infinite-state kernel ϵ-machine—is extracted by a reduced-dimension transform that gives an efficient representation of causal states and their topology. In this way, the system dynamics are represented by a stochastic (ordinary or partial) differential equation that acts on causal states. We introduce an algorithm to estimate the associated evolution operator. Paralleling the Fokker–Planck equation, it efficiently evolves causal-state distributions and makes predictions in the original data space via an RKHS functional mapping. We demonstrate these techniques, together with their predictive abilities, on discrete-time, discrete-value infinite Markov-order processes generated by finite-state hidden Markov models with (i) finite or (ii) uncountably infinite causal states and (iii) continuous-time, continuous-value processes generated by thermally driven chaotic flows. The method robustly estimates causal structure in the presence of varying external and measurement noise levels and for very high-dimensional data.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2026

KRAS4a and KRAS4b show distinct lipid-dependent regulation of RAS-RAF membrane dynamics

KRAS4a and KRAS4b are important regulators of signaling, and their interactions with the plasma membrane are dynamic and influenced by lipid composition. KRAS 4a and 4b have nearly identical globular domains but differ in their membrane-associated hyper variable region (HVR). The functional distinctions between these isoforms remain unclear, particularly with regards to their dependence on specific lipids and the membrane environment. Previous work showed that the membrane orientation of KRAS4b affects its ability to bind to RAF kinase RBDCRD and that the KRAS–RBDCRD complex adopts different poses on the membrane as well as influences the size and composition of the lipid environment. To model differences between KRAS 4a and 4b protein–lipid interactions, we extended the Multiscale Machine-Learned Modeling Infrastructure (MuMMI) to incorporate continuum simulations in the grand canonical ensemble, enabling sampling across macroscopic, coarse-grained, and all-atom resolutions. Using this framework, we systematically altered PIP2 concentrations, KRAS 4a versus 4b, and RAF RBDCRD complexation to assess impacts on membrane–protein interactions and dynamics. Our results reveal that reducing PIP2 shifts and broadens the membrane orientational preference of both KRAS 4b and 4a, with stronger effects on 4b HVR localization versus 4a. We demonstrate that with depletion of the strong negatively charged PIP2 lipid, the less charged phosphatidylserine replaces PIP2. Our findings highlight similarities and distinctions in the dynamics and lipid dependency of KRAS isoforms and suggest that ordering of the local lipid composition by HVRs is a shared property and key modulator of RAS-mediated signaling at the plasma membrane.

Biological and medical sciences↗
DOE OSTI2023

Cummins R-SOFC System Development

The overall purpose of this project was to reduce the Reversible-Solid Oxide Fuel Cell (R-SOFC) system cost by developing two technologies, an improved cell design and the incorporation of an ejector in the fuel recycle loop instead of a blower. A Simulink model of the baseline SOFC system was developed and calibrated with experimental test data. The R-SOFC system model was built by integrating GT Suite developed models of the steam generation components into the baseline Simulink SOFC system model. The ability to run the stack in SOEC operating mode was also added to the model. The system model was used to explore the ability of the R-SOFC system to meet operational constraints on Steam/Carbon ratio and H2 concentration on the fuel side electrode. A CFD ejector model was developed and used to explore a range of ejector design parameters, leading to the final ejector design that was prototyped for testing. A prototype steam ejector was first tested in a laboratory environment using room temperature air. The steam ejector was subsequently tested using the full hot recycle loop with all relevant heat exchangers and steam generation components. The test conditions utilized temperatures, pressures, and flow rates expected in an R-SOFC application. Throughout the experimental testing work, ejector performance test data was used to improve and then validate the CFD ejector model. A CFD cell model was developed and used to optimize thermal gradients, voltage, and cost of a new cell substrate design. A CFD comparison of co-flow and cross-flow cell designs informed the decision to use a co-flow design for the new cell substrate. Multiple rounds of CFD simulation were used to improve the cell design to minimize the variation in air and fuel distribution across cell channels and to minimize the variation in air and fuel distribution across different cells in the stack. A few prototypes of the new cell substrate design were produced and validated in a laboratory environment by thermally spraying and verifying that they met established manufacturing specifications. The cell manufacturing process was adjusted in order to bring these metrics within acceptable tolerances. Cummins’ internal calculations show that the new cell design reduces cost ~50% compared to the baseline cell, while the ejector + superheater/boiler concept reduces cost of the recycle loop by ~40%. The impact of these cost reductions on the cost of producing H2 will depend on the specific system where they are applied. Therefore, a Techno-Economic Analysis was completed using system cost as a variable, and showing how the NREL Current and Future system costs translate into H2 production cost.

Henrichsen, Lars↗
DOE OSTI2023

Quantifying Charge Carrier Localization in PBTTT Using Thermoelectric and Spectroscopic Techniques

Chemically doped poly[2,5-bis(3-alkylthiophen-2-yl)thieno[3,2- b ]thiophene] (PBTTT) shows promise for many organic electronic applications, but rationalizing its charge transport properties is challenging because conjugated polymers are inhomogeneous, with convoluted optical and solid-state transport properties. Herein, we use the semilocalized transport (SLoT) model to quantify how the charge transport properties of PBTTT change as a function of iron(III) chloride (FeCl 3 ) doping level. We use the SLoT model to calculate fundamental transport parameters, including the carrier density needed for metal-like electrical conductivities and the position of the Fermi energy level with respect to the transport edge. We then contextualize these parameters with other polymer-dopant systems and previous PBTTT reports. Additionally, we use grazing incidence wide-angle X-ray scattering and spectroscopic ellipsometry techniques to better characterize inhomogeneity in PBTTT. Our analyses indicate that PBTTT obtains high electrical conductivities due to its quickly rising reduced Fermi energy level, and this rise is afforded by its locally high carrier densities in highly ordered microdomains. Ultimately, this report sets a benchmark for comparing transport properties across polymer-dopant-processing systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Revisiting particle dry deposition and its role in radiative effect estimates

Wet and dry deposition remove aerosols from the atmosphere, and these processes control aerosol lifetime and thus impact climate and air quality. Dry deposition is a significant source of aerosol uncertainty in global chemical transport and climate models. Dry deposition parameterizations in most global models were developed when few particle deposition measurements were available. However, new measurement techniques have enabled more size-resolved particle flux observations. We combined literature measurements with data that we collected over a grassland in Oklahoma and a pine forest in Colorado to develop a dry deposition parameterization. We find that relative to observations, previous parameterizations overestimated deposition of the accumulation and Aitken mode particles, and underestimated in the coarse mode. These systematic differences in observed and modeled accumulation mode particle deposition velocities are as large as an order of magnitude over terrestrial ecosystems. As accumulation mode particles form most of the cloud condensation nuclei (CCN) that influence the indirect radiative effect, this model-measurement discrepancy in dry deposition alters modeled CCN and radiative forcing. We present a revised observationally driven parameterization for regional and global aerosol models. Using this revised dry deposition scheme in the Goddard Earth Observing System (GEOS)-Chem chemical transport model, we find that global surface accumulation-mode number concentrations increase by 62% and enhance the global combined anthropogenic and natural aerosol indirect effect by −0.63 W m −2 . Our observationally constrained approach should reduce the uncertainty of particle dry deposition in global chemical transport models.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Application of a Physics-Informed Convolutional Neural Network for Monitoring the Temperature Fields in High-Temperature Gas Reactors

Here, this work presents current advances in applying a physics-informed convolutional neural network (CNN) to evaluate temperature distributions in advanced reactors. Our goal is to demonstrate that the CNN can reconstruct temperature fields within the solid region of a prismatic fuel assembly in a high-temperature gas reactor (HTGR) with sensor data available in only a few cooling channels. Before that, we showcase the superior performance of the physics-informed CNN in comparison to a purely data-driven multilayer perceptron (MLP), considering a canonical heated channel setup. This analysis shows the advantages of our approach and justifies its choice. The datasets employed here are obtained upon numerical simulations performed with codes under the Nuclear Energy Advanced Modeling and Simulation program. This work is important, as industry experience indicates that the assembly material in HTGR concepts is prone to large thermal-mechanical loads nearing operational limits. This makes it crucial to characterize peak temperatures and their distributions near hot spots. Modern thermocouples are unreliable in these types of harsh environments because of the high neutron fluxes and elevated temperatures involved. The CNN-based field reconstruction represents an attractive solution, enabling sensor arrays in less aggressive locations and augmenting indirect predictions for less accessible regions. The results show that the CNN reduces prediction errors by orders of magnitude in comparison to the MLP, considering the simple yet well-representative heated channel case. In the case of the HTGR fuel assembly, the CNN can successfully reconstruct temperature fields over various cooling regimes. Furthermore, we also explore the algorithm’s ability to detect abnormalities. Interestingly, the CNN proves it has the capacity to detect blockage in one of the noninstrumented cooling channels.

Machine learning↗