SEARCH · Engineering Papers

Results for “high performance space computing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 703 records · Page 39

All sources NASA NTRS DOE OSTI

DOE OSTI2020

Multi-fidelity machine-learning with uncertainty quantification and Bayesian optimization for materials design: Application to ternary random alloys

We present a scale-bridging approach based on a multi-fidelity (MF) machine-learning (ML) framework leveraging Gaussian processes (GP) to fuse atomistic computational model predictions across multiple levels of fidelity. Through the posterior variance of the MFGP, our framework naturally enables uncertainty quantification, providing estimates of confidence in the predictions. We used density functional theory as high-fidelity prediction, while a ML interatomic potential is used as low-fidelity prediction. Practical materials’ design efficiency is demonstrated by reproducing the ternary composition dependence of a quantity of interest (bulk modulus) across the full aluminum–niobium–titanium ternary random alloy composition space. The MFGP is then coupled to a Bayesian optimization procedure, and the computational efficiency of this approach is demonstrated by performing an on-the-fly search for the global optimum of bulk modulus in the ternary composition space. The framework presented in this manuscript is the first application of MFGP to atomistic materials simulations fusing predictions between density functional theory and classical interatomic potential calculations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

NASA NTRS1993

Analysis of data from the energetic gamma-ray experiment on the gamma ray observatory

The work under the Grant has involved continued participation with the Compton Gamma Ray Observatory Energetic Gamma-Ray Experiment Telescope (EGRET) Team in the analysis of data obtained during instrument operations and the preparation of scientific papers and proposals for future observations. The EGRET team was also successful on many proposals for the Phase 3 portion of the mission, including long high galactic latitude studies of the diffuse extragalactic radiation. These studies will be used in a effort to establish whether this radiation is truly diffuse or the sum of radiation from unresolved discrete sources such as radio-loud quasars. The effort involved working remotely by internet connection on the Goddard Space Flight Center Computers where the EGRET data are archived. Students have monitored instrument performance and analyzed data remotely and will continue to do so. The PI has completed the detailed analysis of five viewing periods to search for point sources and this work has been used in developing the first EGRET catalog of sources, soon to be released.

Kniffen, Donald A.↗

NASA NTRS1999

Coherent Doppler Lidar Data Products from Space-Based Platforms

Coherent Doppler lidar is a promising technique for the global measurements of winds using a space-based platform. Doppler lidar produces estimates of the radial component of the velocity vector averaged over the resolution volume of the measurement. Profiles of the horizontal vector winds are produced by scanning the lidar beam or stepping the lidar beam through a sequence of different angles (step-stare). The first design for space-based measurements proposed a conical scan which requires a high power laser to produce acceptable signal levels for every laser pulse. Performance is improved by fixing the laser beam and accumulating the signal from many lidar pulses for each range-gate. This also improves the spatial averaging of the wind estimates and reduces the threshold signal energy required for a good estimate. Coherent Doppler lidar performance for space-based operation is determined using computer simulations and including the wind variability over the measurement volume as well as the variations of the atmospheric aerosol backscatter.

Frehlich, Rod↗

NASA NTRS2003

Robust Airfoil Optimization in High Resolution Design Space

The robust airfoil shape optimization is a direct method for drag reduction over a given range of operating conditions and has three advantages: (1) it prevents severe degradation in the off-design performance by using a smart descent direction in each optimization iteration, (2) it uses a large number of B-spline control points as design variables yet the resulting airfoil shape is fairly smooth, and (3) it allows the user to make a trade-off between the level of optimization and the amount of computing time consumed. The robust optimization method is demonstrated by solving a lift-constrained drag minimization problem for a two-dimensional airfoil in viscous flow with a large number of geometric design variables. Our experience with robust optimization indicates that our strategy produces reasonable airfoil shapes that are similar to the original airfoils, but these new shapes provide drag reduction over the specified range of Mach numbers. We have tested this strategy on a number of advanced airfoil models produced by knowledgeable aerodynamic design team members and found that our strategy produces airfoils better or equal to any designs produced by traditional design methods.

Li, Wu↗

NASA NTRSDate not supplied

Enabling Open and Interoperable Science: Multi-Omics Data Processing Platform with NASA GeneLab Standardized Bioinformatics Workflows for Space and Earth Research

Multi-omics biological data continues to be generated at an astounding pace. Genomics, transcriptomics, metabolomics, and proteomics, or collectively known as multi-omics data, are used to assess biological functions, and provide invaluable insights into human, animal, plant, and environmental health both on Earth and in Space. Despite the abundance of these valuable data, the need for bioinformatics expertise, particularly as it relates to the niche filed of space biology, and a lack of accessible resources for processing these data limit their usefulness in deriving biological insights. The NASA Open Science Data Repository (OSDR) provides access to omics data from various spaceflight and analog studies. To enhance the accessibility and reusability of these data, GeneLab (part of OSDR) designs and implements standardized, community-driven, open-source bioinformatics workflows to transform raw omics data into standardized processed data. Currently, GeneLab-processed data from hundreds of space studies have been reused for meta-analyses. This has led to new insights and scientific publications that extend beyond the initial research, thereby enriching our understanding of molecular-scale biological responses to the space environment. To make these bioinformatics workflows open and accessible, GeneLab teamed up with DOE-funded initiatives, including the National Microbiome Data Collaborative (NMDC), to create the NASA EDGE [Empowering the Development of Genomics Expertise] Bioinformatics web-based platform. NASA EDGE utilizes shared compute resources to run the GeneLab standardized bioinformatics workflows, which eliminates the need for researchers to have their own high performance computing cluster. The web-based platform makes complicated biological analyses incredibly easy to perform, thus expanding the reach of these analyses to bioinformatics novices, students, and even citizen scientists enabling them to contribute to scientific discoveries and progress. The authors will demonstrate how the NASA EDGE platform can be used to process microbial omics data hosted on OSDR as well as user-generated omics datasets using GeneLab’s standard workflows.

Amanda M. Saravia-Butler↗

DOE OSTI2023

Optimizing shift selection in multilevel Monte Carlo for disconnected diagrams in lattice QCD

The calculation of disconnected diagram contributions to physical signals is a computationally expensive task in Lattice QCD. To extract the physical signal, the trace of the inverse Lattice Dirac operator, a large sparse matrix, must be stochastically estimated. Because the variance of the stochastic estimator is typically large, variance reduction techniques must be employed. Multilevel Monte Carlo (MLMC) methods reduce the variance of the trace estimator by utilizing a telescoping sequence of estimators. Frequency Splitting is one such method that uses a sequence of inverses of shifted operators to estimate the trace of the inverse lattice Dirac operator, however there is no a priori way to select the shifts that minimize the cost of the multilevel trace estimation. Here we present a sampling and interpolation scheme that is able to predict the variances associated with Frequency Splitting under displacements of the underlying space time lattice. The interpolation scheme is able to predict the variances to high accuracy and therefore chooses shifts that correspond to an approximate minimum of the cost for the trace estimation. We show that Frequency Splitting with the chosen shifts displays significant speedups over multigrid deflation, and that these shifts can be used for multiple configurations within the same ensemble with no penalty to performance.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2012

Active Mirror Predictive and Requirements Verification Software (AMP-ReVS)

This software is designed to predict large active mirror performance at various stages in the fabrication lifecycle of the mirror. It was developed for 1-meter class powered mirrors for astronomical purposes, but is extensible to other geometries. The package accepts finite element model (FEM) inputs and laboratory measured data for large optical-quality mirrors with active figure control. It computes phenomenological contributions to the surface figure error using several built-in optimization techniques. These phenomena include stresses induced in the mirror by the manufacturing process and the support structure, the test procedure, high spatial frequency errors introduced by the polishing process, and other process-dependent deleterious effects due to light-weighting of the mirror. Then, depending on the maturity of the mirror, it either predicts the best surface figure error that the mirror will attain, or it verifies that the requirements for the error sources have been met once the best surface figure error has been measured. The unique feature of this software is that it ties together physical phenomenology with wavefront sensing and control techniques and various optimization methods including convex optimization, Kalman filtering, and quadratic programming to both generate predictive models and to do requirements verification. This software combines three distinct disciplines: wavefront control, predictive models based on FEM, and requirements verification using measured data in a robust, reusable code that is applicable to any large optics for ground and space telescopes. The software also includes state-of-the-art wavefront control algorithms that allow closed-loop performance to be computed. It allows for quantitative trade studies to be performed for optical systems engineering, including computing the best surface figure error under various testing and operating conditions. After the mirror manufacturing process and testing have been completed, the software package can be used to verify that the underlying requirements have been met.

Basinger, Scott A.↗

DOE OSTI2025

Breaking the mold: Overcoming the time constraints of molecular dynamics on general-purpose hardware

The evolution of molecular dynamics (MD) simulations has been intimately linked to that of computing hardware. For decades following the creation of MD, simulations have improved with computing power along the three principal dimensions of accuracy, atom count (spatial scale), and duration (temporal scale). Since the mid-2000s, computer platforms have, however, failed to provide strong scaling for MD, as scale-out central processing unit (CPU) and graphics processing unit (GPU) platforms that provide substantial increases to spatial scale do not lead to proportional increases in temporal scale. Important scientific problems therefore remained inaccessible to direct simulation, prompting the development of increasingly sophisticated algorithms that present significant complexity, accuracy, and efficiency challenges. While bespoke MD-only hardware solutions have provided a path to longer timescales for specific physical systems, their impact on the broader community has been mitigated by their limited adaptability to new methods and potentials. In this work, we show that a novel computing architecture, the Cerebras wafer scale engine, completely alters the scaling path by delivering unprecedentedly high simulation rates up to 1.144 M steps/s for 200 000 atoms whose interactions are described by an embedded atom method potential. This enables direct simulations of the evolution of materials using general-purpose programmable hardware over millisecond timescales, dramatically increasing the space of direct MD simulations that can be carried out. In this paper, we provide an overview of advances in MD over the last 60 years and present our recent result in the context of historical MD performance trends.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2021

Enhanced Data Efficiency Using Deep Neural Networks and Gaussian Processes for Aerodynamic Design Optimization

Adjoint-based optimization methods are attractive for aerodynamic shape design primarily due to their computational costs being independent of the dimensionality of the input space and their ability to generate high-fidelity gradients that can then be used in a gradient-based optimizer. This makes them very well suited for high-fidelity simulation based aerodynamic shape optimization of highly parametrized geometries such as aircraft wings. However, the development of adjoint-based solvers involve careful mathematical treatment and their implementation require detailed software development. Furthermore, they can become prohibitively expensive when multiple optimization problems are being solved, each requiring multiple restarts to circumvent local optima. In this work, we propose a machine learning enabled, surrogate-based framework that replaces the expensive adjoint solver, without compromising on predicting predictive accuracy. Specifically, we first train a deep neural network (DNN) from training data generated from evaluating the high-fidelity simulation model on a model-agnostic design of experiments on the geometry shape parameters. The optimum shape may then be computed by using a gradient-based optimizer coupled with the trained DNN. Subsequently, we also perform a gradient-free Bayesian optimization, where the trained DNN is used as the prior mean. We observe that the latter framework (DNN-BO) improves upon the DNN-only based optimization strategy for the same computational cost. Overall, this framework predicts the true optimum with very high accuracy, while requiring far fewer high-fidelity function calls compared to the adjoint-based method. Furthermore, we show that multiple optimization problems can be solved with the same machine learning model with high accuracy, to amortize the offline costs associated with constructing our models. Our methodology finds applications in the early stages of aerospace design. (C) 2021 Published by Elsevier Masson SAS.

Renganathan, S. Ashwin↗

NASA NTRSDate not supplied

Computational and Physics-Based Modeling for the Development of in-Space Welding Technology

Welding will be an enabling technology for in-space assembly and manufacturing (ISAM) and in-space repair operations in support of the growing Space Economy. Prospective use cases of welding include metal repair on long-duration lunar surface missions and assembly of large structures that exceed dimensions of the currently available launch vehicle envelopes such as space arrays, tall lunar transmission towers, or even large transit vehicles powered by nuclear propulsion. There is a brief history of beam welding for in-space use. Soviet and NASA astronauts performed electron beam welds (EBW) in space in the late-1960s through mid-1980s and several parabolic flights of laser beam welding (LBW) were undertaken in the USA and Canada in the early-1990s. Work has since stagnated until the recent boom of the in-space economy has demonstrated need for continued development of EBW and LBW processes, which are desirable for their compactness, low mass, low power consumption, and high wall plug efficiency. No space-based manufacturing laboratory capable of demonstrating, developing, and applying welding processes exists, and costs are high to develop and fly individual space-based trials. This drives the current approach to develop welding technology with computational and physics-based models in tandem with earth-based experiments. Several approaches are taken to complement and drive the selection of instrumentation for planned parabolic and suborbital flights. First, a thermal model of Skylab EBW experiments is established and compared to empirical data. Next, the model is used to show variation between Earth-based and in-space weld structure and properties through exploration of key variables that define the Low earth orbit and lunar surface space environments, namely temperature (40 – 400 K), pressure (10-10 to 104 Pa), and gravity (10-6 g to 1 g). Finally, physical processes associated with heat and mass transport during welding are evaluated through several case studies across the above variable ranges demonstrating need for further research.

in-space welding↗

NASA NTRS2014

HRP's Healthcare Spin-Offs Through Computational Modeling and Simulation Practice Methodologies

Spaceflight missions expose astronauts to novel operational and environmental conditions that pose health risks that are currently not well understood, and perhaps unanticipated. Furthermore, given the limited number of humans that have flown in long duration missions and beyond low Earth-orbit, the amount of research and clinical data necessary to predict and mitigate these health and performance risks are limited. Consequently, NASA's Human Research Program (HRP) conducts research and develops advanced methods and tools to predict, assess, and mitigate potential hazards to the health of astronauts. In this light, NASA has explored the possibility of leveraging computational modeling since the 1970s as a means to elucidate the physiologic risks of spaceflight and develop countermeasures. Since that time, substantial progress has been realized in this arena through a number of HRP funded activates such as the Digital Astronaut Project (DAP) and the Integrated Medical Model (IMM). Much of this success can be attributed to HRP's endeavor to establish rigorous verification, validation, and credibility (VV&C) processes that ensure computational models and simulations (M&S) are sufficiently credible to address issues within their intended scope. This presentation summarizes HRP's activities in credibility of modeling and simulation, in particular through its outreach to the community of modeling and simulation practitioners. METHODS: The HRP requires all M&S that can have moderate to high impact on crew health or mission success must be vetted in accordance to NASA Standard for Models and Simulations, NASA-STD-7009 (7009) [5]. As this standard mostly focuses on engineering systems, the IMM and DAP have invested substantial efforts to adapt the processes established in this standard for their application to biological M&S, which is more prevalent in human health and performance (HHP) and space biomedical research and operations [6,7]. These methods have also generated substantial interest by the broader medical community though institutions like the National Institutes of Health (NIH) and the Food and Drug Administration (FDA) to develop similar standards and guidelines applicable to the larger medical operations and research community. DISCUSSION: Similar to NASA, many leading government agencies, health institutions and medical product developers around the world are recognizing the potential of computational M&S to support clinical research and decision making. In this light, substantial investments are being made in computational medicine and notable discoveries are being realized [8]. However, there is a lack of broadly applicable practice guidance for the development and implementation of M&S in clinical care and research in a manner that instills confidence among medical practitioners and biological researchers [9,10]. In this presentation, we will give an overview on how HRP is working with the NIH's Interagency Modeling and Analysis Group (IMAG), the FDA and the American Society of Mechanical Engineers (ASME) to leverage NASA's biomedical VV&C processes to establish a new regulatory standard for Verification and Validation in Computational Modeling of Medical Devices, and Guidelines for Credible Practice of Computational Modeling and Simulation in Healthcare.

Mulugeta, Lealem↗

DOE OSTI2020

Time-stepping DPG formulations for the heat equation

For a wide range of PDEs, the discontinuous Petrov–Galerkin (DPG) methodology of Demkowicz and Gopalakrishnan provides discrete stability starting from a coarse mesh and minimization of the residual in a user-controlled norm, among other appealing features. Research on DPG for transient problems has mainly focused on spacetime discretizations, which has theoretical advantages, but practical costs for computations and software implementations. The sole examination of time-stepping DPG formulations was performed by Führer, Heuer, and Gupta, who applied Rothe’s method to an ultraweak formulation of the heat equation to develop an implicit time-stepping scheme; their work emphasized theoretical results, including error estimates in time and space. Here, we follow Führer, Heuer, and Gupta in examining the heat equation; our focus is on numerical experiments, examining the stability and accuracy of several formulations, including primal as well as ultraweak, and explicit as well as implicit and Crank–Nicolson time-stepping schemes. We are additionally interested in communication-avoiding algorithms, and we therefore include a highly experimental formulation that places all the trace terms on the right-hand side of the equation.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

An extended Vlasov–Fokker–Planck approach for kinetic simulations of laser plasmas

Vlasov–Fokker–Planck simulation codes occupy an important niche in modeling laser-produced plasmas, since they are well suited to studying the effect of collisions on electron kinetic phenomena, especially energy transport. One of the most important elements of energy transport is the absorption of laser light by the plasma; however, simulating this in detail requires resolving oscillations of the laser light, whose characteristic timescale is orders of magnitude shorter than the simulation time needed to study transport physics. For this reason, most Vlasov–Fokker–Planck codes used to study electron transport in laser plasmas rely on simplified models of the laser–plasma coupling. Their underlying assumptions nominally preclude their use for modeling laser light having short-scale structure in space or time, such as broadband lasers. In this work, we derive a more general computational framework suitable for arbitrarily structured laser fields. Furthermore, our approach is based on an extended set of Vlasov–Fokker–Planck equations that separately solve for the low- and high-frequency plasma response. We implement these extended Vlasov–Fokker–Planck equations in the spherical harmonic code K2 and demonstrate the performance of the method on several laser absorption test problems, with particular attention to the judicious selection of time steps, time integrators, and spherical harmonic truncation, according to the intensity and spectrum of the laser light under consideration. Comparison with the widely used Langdon absorption operator shows the Langdon operator performs remarkably well for predicting laser heating in the simple cases considered here, even in situations that would seem to violate its underlying assumptions.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

NASA NTRS2011

CFD Study of the J2-X LOX Two-Bladed Inducer in the Presence of a Circumferential Groove

Results are presented of a computational fluid dynamics (CFD) study done in support of water flow experiments of the J-2X Oxidizer Turbopump (OTP) 2-bladed inducer with a circumferential groove that were conducted at Marshall Space Flight Center (MSFC). Sub-scale water flow testing results indicate that the circumferential groove greatly reduces synchronous cavitation and subsequent bearing loads at a minimal performance cost, but the energy reappears as high order cavitation (HOC) that spans a relatively large frequency range. Thus, HOC may have implications for the full-scale OTP inducer in terms of reduced structural margin at higher mode frequencies. Simulations using the LOCI-Stream CFD program were conducted in order to explore the root physical cause of the HOC. It was found that the axial recirculation pattern in the circumferential groove causes high-swirl fluid to interact with the nearly-axial incoming fluid just above the inducer blades. The high-shear interface between the fluids is Kelvin-Helmholtz unstable, resulting in trains of low pressure regions or 'pearls' forming near the upstream edge of the groove. When the pressure in these regions becomes low enough and they get cut by the blade leading edge, HOC is thought to occur. Although further work is required, the numerical models indicate that the root cause of HOC is hydrodynamic. That is, the pearls are always present, even when cavitation is not. Comparisons to ongoing water flow experiments will be discussed, as well as predictions for the full-scale OTP inducer.

Coker, Robert↗

NASA NTRS2013

Project Morpheus: Morpheus 1.5A Lander Failure Investigation Results

On August 9, 2012 the Morpheus 1.5A vehicle crashed shortly after lift off from the Kennedy Space Center. The loss was limited to the vehicle itself which was pre-declared to be a test failure and not a mishap. The Morpheus project is demonstrating advanced technologies for in space and planetary surface vehicles including: autonomous flight control, landing site hazard identification and safe site selection, relative surface and hazard navigation, precision landing, modular reusable flight software, and high performance, non-toxic, cryogenic liquid Oxygen and liquid Methane integrated main engine and attitude control propulsion system. A comprehensive failure investigation isolated the fault to the Inertial Measurement Unit (IMU) data path to the flight computer. Several improvements have been identified and implemented for the 1.5B and 1.5C vehicles.

Devolites, Jennifer L.↗

DOE OSTI2025

Multi-Objective Optimization of Uranium Target Assembly–3: A Comparison of Genetic and Traditional Methods

Commonly produced as a byproduct of uranium fission, 99 Mo is a key medical isotope that is in high demand in the United States. An international goal is to switch from medical isotope production technologies that require highly enriched uranium to medical isotope production technologies that require only low-enriched uranium. Niowave Inc. is contributing to this goal by developing an accelerator-driven subcritical assembly called the Uranium Target Assembly (UTA). This work compares the performance of Dakota’s Multi-Objective Genetic Algorithm (MOGA) against traditional sensitivity analysis in the neutronic optimization of the UTA-3 system. The design objectives are k-eigenvalue (k eff ) and natural uranium fission power, which are directly correlated with the amount of 99 Mo produced. Dakota:MOGA did not perform as well as human engineering ingenuity in optimization studies with high numbers of input parameters, such as fuel rod type selection and fuel rod placement. However, Dakota:MOGA did outperform traditional sensitivity analysis in optimization studies with fewer than 20 parameters and revealed the degree to which each parameter influences the optimal design space for k eff and natural uranium fission power (to a lesser extent). As the design model became more complex in the final stage of design, the computational resources required to calculate the design objective values in the Monte Carlo N-Particle transport code from selected input parameter combinations limited Dakota:MOGA’s performance, and, unfortunately, human intervention was required to discern the optimal design space. In conclusion, future work will attempt to reduce computational resource constraints by incorporating areduced-order neutronics model into the optimization cycle.

Accelerator-driven systems↗

DOE OSTI2022

Inverse Analysis with Variational Autoencoders: A Comparison of Shallow and Deep Networks

Inverse problems are applied to determine unknown properties by matching observational data with a physical model that often takes many parameters as input. To overcome the underconstrained nature of inverse problems and achieve good performance, an approach is presented involving regularization with a technique known as a variational autoencoder (VAE), which is trained to map a high-dimensional parameter space with a complex structure to a low-dimensional latent space with a simple structure. We apply this approach to unconditioned realizations of the parameters (heterogeneous hydraulic fields) for a hydrogeological inverse problem. Two types of hydraulic conductivity fields are used to evaluate the characterization for the different levels of heterogeneity complexity of the physical inputs. This approach keeps the computational cost of generating the training data low. The reason is unconditioned realizations neither rely on the observational data used to perform the inverse analysis nor require any groundwater flow model forward runs. In addition, this approach applies regularization on a low-dimensional latent space from the VAE and increases optimization efficiency through automatic differentiation. Furthermore, two different neural network (NN) structures are tested for their utility in using the VAE for inverse analysis. The performance of a deep, convolutional neural network strongly depends on the dimensionality of the latent space, which requires tuning. In contrast, a shallow, dense neural network provides consistently accurate characterization without tuning. Furthermore, our approach evaluates the advantages of the shallow, dense neural network over the deep, convolutional one and enables future application to a wide range of inverse problems.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Fusion RF Modeling Machine Learning (FusionML_RF) v1.0

FusionML_RF consists of multiple codes and trained machine learning (ML) models that perform low-cost output modeling from the Genray-CQL3D. Three machine learning techniques (multilayer perceptron, random forest, and Gaussian process) provide fast surrogate models for lower hybrid current drive (LHCD) simulations. For example, completing a single GENRAY/CQL3D simulation without radial diffusion of fast electrons requires several minutes of wall-clock time. On the other hand, these ML models achieve ~ms of inference time with high accuracy across the input parameter space. This software collection consists of multiple components. (1) codes that use ML methods and precomputed Genray-CQL3D simulation output to build regression models that enable approximate computations of Genray-CLQ3D outputs from arbitrary but physically meaningful input parameters (surrogate modeling); (2) three trained models created by the team, using a database of 16,000+ GENRAY/CQL3D simulations, to study the performance of ML models for surrogate modeling; (3) codes that load the trained models and simulation data, and then compute mean squared error between the models' predictions and the ground truth of simulation output data. This collection is being made available in conjunction with a scientific publication about the work to promote reusability and provide an artifact of the scientific work.

Bai, Zhe↗