Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data reduction techniques”

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 199 records · Page 11

NEAR-INFRARED THERMAL EMISSION DETECTIONS OF A NUMBER OF HOT JUPITERS AND THE SYSTEMATICS OF GROUND-BASED NEAR-INFRARED PHOTOMETRY

We present detections of the near-infrared thermal emission of three hot Jupiters and one brown dwarf using the Wide-field Infrared Camera (WIRCam) on the Canada-France-Hawaii Telescope (CFHT). These include Ks-band secondary eclipse detections of the hot Jupiters WASP-3b and Qatar-1b and the brown dwarf KELT-1b. We also report Y-band, K {sub CONT}-band, and two new and one reanalyzed Ks-band detections of the thermal emission of the hot Jupiter WASP-12b. We present a new reduction pipeline for CFHT/WIRCam data, which is optimized for high precision photometry. We also describe novel techniques for constraining systematic errors in ground-based near-infrared photometry, so as to return reliable secondary eclipse depths and uncertainties. We discuss the noise properties of our ground-based photometry for wavelengths spanning the near-infrared (the YJHK bands), for faint and bright stars, and for the same object on several occasions. For the hot Jupiters WASP-3b and WASP-12b we demonstrate the repeatability of our eclipse depth measurements in the Ks band; we therefore place stringent limits on the systematics of ground-based, near-infrared photometry, and also rule out violent weather changes in the deep, high pressure atmospheres of these two hot Jupiters at the epochs of our observations.

79 ASTRONOMY AND ASTROPHYSICS↗

A probabilistic graphical model based stochastic input model construction

Model reduction techniques have been widely used in modeling of high-dimensional stochastic input in uncertainty quantification tasks. However, the probabilistic modeling of random variables projected into reduced-order spaces presents a number of computational challenges. Due to the curse of dimensionality, the underlying dependence relationships between these random variables are difficult to capture. In this work, a probabilistic graphical model based approach is employed to learn the dependence by running a number of conditional independence tests using observation data. Thus a probabilistic model of the joint PDF is obtained and the PDF is factorized into a set of conditional distributions based on the dependence structure of the variables. The estimation of the joint PDF from data is then transformed to estimating conditional distributions under reduced dimensions. To improve the computational efficiency, a polynomial chaos expansion is further applied to represent the random field in terms of a set of standard random variables. This technique is combined with both linear and nonlinear model reduction methods. Numerical examples are presented to demonstrate the accuracy and efficiency of the probabilistic graphical model based stochastic input models. - Highlights: • Data-driven stochastic input models without the assumption of independence of the reduced random variables. • The problem is transformed to a Bayesian network structure learning problem. • Examples are given in flows in random media.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

acdc – Automated Contamination Detection and Confidence estimation for single-cell genome data

Abstract Background A major obstacle in single-cell sequencing is sample contamination with foreign DNA. To guarantee clean genome assemblies and to prevent the introduction of contamination into public databases, considerable quality control efforts are put into post-sequencing analysis. Contamination screening generally relies on reference-based methods such as database alignment or marker gene search, which limits the set of detectable contaminants to organisms with closely related reference species. As genomic coverage in the tree of life is highly fragmented, there is an urgent need for a reference-free methodology for contaminant identification in sequence data. Results We present acdc, a tool specifically developed to aid the quality control process of genomic sequence data. By combining supervised and unsupervised methods, it reliably detects both known and de novo contaminants. First, 16S rRNA gene prediction and the inclusion of ultrafast exact alignment techniques allow sequence classification using existing knowledge from databases. Second, reference-free inspection is enabled by the use of state-of-the-art machine learning techniques that include fast, non-linear dimensionality reduction of oligonucleotide signatures and subsequent clustering algorithms that automatically estimate the number of clusters. The latter also enables the removal of any contaminant, yielding a clean sample. Furthermore, given the data complexity and the ill-posedness of clustering, acdc employs bootstrapping techniques to provide statistically profound confidence values. Tested on a large number of samples from diverse sequencing projects, our software is able to quickly and accurately identify contamination. Results are displayed in an interactive user interface. Acdc can be run from the web as well as a dedicated command line application, which allows easy integration into large sequencing project analysis workflows. Conclusions Acdc can reliably detect contamination in single-cell genome data. In addition to database-driven detection, it complements existing tools by its unsupervised techniques, which allow for the detection of de novo contaminants. Our contribution has the potential to drastically reduce the amount of resources put into these processes, particularly in the context of limited availability of reference species. As single-cell genome data continues to grow rapidly, acdc adds to the toolkit of crucial quality assurance tools.

59 BASIC BIOLOGICAL SCIENCES↗

Speed and Memory Improvements to MCNP6 Delayed-Gamma Line Treatment

The delayed-gamma (DG) line-emission (line) sampling was first introduced into MCNPX in 2005. At the time, storage demands for problems with large numbers of DG lines was a major concern. In order to accommodate large amounts of line data while keeping the memory footprint of such a problem from exceeding memory limits, complicated logic was included to sift through various storage options. In mid-2015 concerns arose over the adequacy of the MCNP6 DG source option (inherited from MCNPX) with line sampling for modeling high-fidelity problems. One indication that there was a problem came when studying the MCNP6 produced DG spectra of U-238 and its progeny using high-fidelity spectroscopic tallies when compared to DG line amplitudes directly from ENDF/B-VII.1 emission data. Despite producing a correct integral number of gammas, the comparison to ENDF data showed a broadening in the energy lines and a reduction in the amplitude of more prominent lines. It was later found that the cause of this broadening was one of the memory-saving techniques used to store DG line data when problems required many DG lines. Specifically, the DG lines were integrated into fine energy bins for sampling. When a DG line was to be sampled, a random number would be used to choose which energy bin to sample based on emission probability. Once chosen, a second random number was used to perform nearest-line sampling in that bin. The integration of lines led to an oversampling of low-probability lines, especially when in close proximity to high-probability lines. This treatment also caused a similar issue with the time-dependent behavior of DG line emission which occurred when low-probability lines were sampled near high probability lines with the former decay constant being used. The effect on time-dependent gamma production was demonstrated in analytic benchmarks performed by Weldon et al.. To address these issues a new gamma line storage and sampling treatment was introduced in 2015. This new treatment, stores gamma emission data in a line-by-line CDF which is sampled directly, thereby resolving both the line-broadening issue, and the time-dependence issue as was demonstrated by Tutt el al.. In addition, the new DG sampling treatment was written to use dynamically dimensioned storage arrays as opposed to large fixed-dimensioned arrays, as in the former treatment, and also makes more efficient use of line data that is stored by reducing or eliminating repeated data. As a result, storage arrays are generated only for DG data relevant to the problem being executed. The result should be a reduction in both memory requirements and runtime. However, these reductions could not be realized until the logic for the old treatment could be removed. This paper provides an update on recent work which has focused on the removal of the old DG sampling logic and attempts to quantify the memory reduction and speed up of selected problems.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Deep Learning-based Parameterization of Complex 3D CO2 Saturation Data in Large-scale Geological Carbon Storage

In deep learning (DL), dimension reduction plays a pivotal role in improving training efficiency and minimizing overfitting, especially when working with complex datasets like three-dimensional (3D) saturation data. In the context of geological carbon storage (GCS), 3D saturation data introduces unique challenges due to its sparse nature and sharp transitions at plume boundaries, known as shock fronts. To tackle these challenges, we developed a novel DL framework that combines dimension reduction with advanced 3D reconstruction techniques. Our approach utilizes latent variables derived from 2D average saturation fields to efficiently capture the essential features of high-dimensional data while reducing the number of variables. This enhances both the robustness and accuracy of DL models, making the framework more practical for real-world applications. By offering a tailored solution for modeling complex 3D saturation dynamics, this framework holds significant potential for environmental monitoring, energy storage, and other geological applications.

Wang, Hongsheng [University of Texas at Austin]↗

Towards Effective Clustering Techniques for the Analysis of Electric Power Grids

Clustering is an important data analysis technique with numerous applications in the analysis of electric power grids. Standard clustering techniques are oblivious to the rich structural and dynamic information available for power grids. Therefore, by exploiting the inherent topological and electrical structure in the power grid data, we propose new methods for clustering with applications to model reduction, locational marginal pricing, phasor measurement unit (PMU or synchrophasor) placement, and power system protection. We focus our attention on model reduction for analysis based on time-series information from synchrophasor measurement devices, and spectral techniques for clustering. By comparing different clustering techniques on two instances of realistic power grids we show that the solutions are related and therefore one could leverage that relationship for a computational advantage. Thus, by contrasting different clustering techniques we make a case for exploiting structure inherent in the data with implications for several domains including power systems.

clustering, k-means, power grid, graphs↗

Combining X-ray Absorption and X-ray Diffraction Techniques for in Situ Studies of Chemical Transformations in Heterogeneous Catalysis:Advantages and Limitations

Recent advances in catalysis instrumentations include synchrotron-based facilities where time-resolved X-ray scattering and absorption techniques are combined in the same in situ or operando experiment to study catalysts at work. To evaluate the advances and limitations of this method, we performed a series of experiments at the new XAFS/XRD instrument in the National Synchrotron Light Source. Nearly simultaneous X-ray diffraction (XRD) and X-ray absorption fine-structure (XAFS) measurements of structure and kinetics of several catalysts under reducing or oxidizing conditions have been performed and carefully analyzed. For CuFe{sub 2}O{sub 4} under reducing conditions, the combined use of the two techniques allowed us to obtain accurate data on kinetics of nucleation and growth of metallic Cu. For the inverse catalyst CuO/CeO{sub 2} that underwent isothermal reduction (with CO) and oxidation (with O{sub 2}), the XAFS data measured in the same experiment with XRD revealed strongly disordered Cu species that went undetected by diffraction. These and other examples emphasize the unique sensitivity of these two complementary methods to follow catalytic processes in the broad ranges of length and time scales.

43 PARTICLE ACCELERATORS↗

Rapid detection of rare events from in situ X-ray diffraction data using machine learning

High-energy X-ray diffraction methods can non-destructively map the 3D microstructure and associated attributes of metallic polycrystalline engineering materials in their bulk form. These methods are often combined with external stimuli such as thermo-mechanical loading to take snapshots of the evolving microstructure and attributes over time. However, the extreme data volumes and the high costs of traditional data acquisition and reduction approaches pose a barrier to quickly extracting actionable insights and improving the temporal resolution of these snapshots. This article presents a fully automated technique capable of rapidly detecting the onset of plasticity in high-energy X-ray microscopy data. The technique is computationally faster by at least 50 times than the traditional approaches and works for data sets that are up to nine times sparser than a full data set. This new technique leverages self-supervised image representation learning and clustering to transform massive data sets into compact, semantic-rich representations of visually salient characteristics ( e.g. peak shapes). These characteristics can rapidly indicate anomalous events, such as changes in diffraction peak shapes. It is anticipated that this technique will provide just-in-time actionable information to drive smarter experiments that effectively deploy multi-modal X-ray diffraction methods spanning many decades of length scales.

Zheng, Weijian↗

AIACHNE's contribution for Nuclear Energy Agency Working Party on International Nuclear Data Evaluation Co-operation Subgroup 50

The AIACHNE (AI/ML Informed cAlifornium CHi Nuclear data Experiment) project aims at designing an experiment for the 252 Cf Prompt Fission Neutron Spectrum (PFNS) that explores systematic biases in an experimental database retrieved from the EXFOR databases. To that end, machine learning (ML) methods were applied to pint-point measurement features likely related to bias. From that information, we selected a feature that should be explored by the AIACHNE experiment. Measurement features are metadata encapsulating all pertinent information about the physical measurement and analysis techniques. Examples are, for instance, what neutron and fission detectors were used for the physical metadata, and what background reduction techniques were employed for analysis techniques. Such metadata were retrieved both from EXFOR entries as well as the literature of data sets described in detail in Reference 2 (at the end of the article).

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

AIACHNE's contribution for Nuclear Energy Agency Working Party on International Nuclear Data Evaluation Co-operation Subgroup 50

The AIACHNE (AI/ML Informed cAlifornium CHi Nuclear data Experiment) project aims at designing an experiment for the 252 Cf Prompt Fission Neutron Spectrum (PFNS) that explores systematic biases in an experimental database retrieved from the EXFOR databases. To that end, machine learning (ML) methods were applied to pint-point measurement features likely related to bia. From that information, we selected a feature that should be explored by the AIACHNE experiment. Measurement features are metadata encapsulating all pertinent information about the physical measurement and analysis techniques. Examples are, for instance, what neutron and fission detectors were used for the physical metadata, and what background reduction techniques were employed for analysis techniques. Such metadata were retrieved both from EXFOR entries as well as the literature of data sets described in detail in Ref. [2]. The prerequisite for applying machine learning techniques is casting the metadata into a format that can be parsed by the algorithm. This step might seem trivial but requires to find a unique language where metadata that carry the same physics meaning across several experiments must have the same identifier. One example is, for instance, the neutron detector. As seen in Figure 1, the machine learning code identified the use of 6 Li detectors as being related to bias in some datasets of the AIACHNE 252 Cf PFNS experimental database. In fact, here are several experiments that used neutron detectors containing 6Li in the database, for instance for the example below. EXFOR format has a unique keywords describing detectors such as “SCIN” or “GLASD”. One may think that these keywords are already sufficient descriptors for ML to uniquely find an issue. However, “SCIN” (used for [3, 4]) and “GLASD” (used for [5]) fail to inform the algorithm what is the active material in the detector. And, the key common issue leading to bias in 252 Cf related to neutron detectors is not whether it is a glass detector or a scintillator. No, the issue is that 6 Li was within both detector types and that even small mistakes in the detector response functions around approximately 200 keV are amplified by the 6 Li(n,α) resonance there leading to bias in data as highlighted in Fig. 1 and Ref. [1]. Hence, the features describing the neutron detector must call out the active material in the detector, rather than the existing EXFOR detector keyword, that the ML algorithm can find physically meaningful features related to bias. The AIACHNE team used a precursor of the WPEC (Working Party on International Nuclear Data Evaluation Co-operation) SG(Subgroup)-50 format to store the metadata for the ML analysis.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

On single-crystal total scattering data reduction and correction protocols for analysis in direct space

Data reduction and correction steps and processed data reproducibility in the emerging single-crystal total-scattering-based technique of three-dimensional differential atomic pair distribution function (3D-ΔPDF) analysis are explored. All steps from sample measurement to data processing are outlined using a crystal of CuIr 2 S 4 as an example, studied in a setup equipped with a high-energy X-ray beam and a flat-panel area detector. Computational overhead as pertains to data sampling and the associated data-processing steps is also discussed. Various aspects of the final 3D-ΔPDF reproducibility are explicitly tested by varying the data-processing order and included steps, and by carrying out a crystal-to-crystal data comparison. Situations in which the 3D-ΔPDF is robust are identified, and caution against a few particular cases which can lead to inconsistent 3D-ΔPDFs is noted. Although not all the approaches applied herein will be valid across all systems, and a more in-depth analysis of some of the effects of the data-processing steps may still needed, the methods collected herein represent the start of a more systematic discussion about data processing and corrections in this field.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Exploring the limits of waveform correlation event detection as applied to three earthquake aftershock sequences.

Swarms of earthquakes and/or aftershock sequences can dramatically increase the level of seismicity in a region for a period of time lasting from days to months, depending on the swarm or sequence. Such occurrences can provide a large amount of useful information to seismologists. For those who monitor seismic events for possible nuclear explosions, however, these swarms/sequences are a nuisance. In an explosion monitoring system, each event must be treated as a possible nuclear test until it can be proven, to a high degree of confidence, not to be. Seismic events recorded by the same station with highly correlated waveforms almost certainly have a similar location and source type, so clusters of events within a swarm can quickly be identified as earthquakes. We have developed a number of tools that can be used to exploit the high degree of waveform similarity expected to be associated with swarms/sequences. Dendro Tool measures correlations between known events. The Waveform Correlation Detector is intended to act as a detector, finding events in raw data which correlate with known events. The Self Scanner is used to find all correlated segments within a raw data steam and does not require an event library. All three techniques together provide an opportunity to study the similarities of events in an aftershock sequence in different ways. To comprehensively characterize the benefits and limits of waveform correlation techniques, we studied 3 aftershock sequences, using our 3 tools, at multiple stations. We explored the effects of station distance and event magnitudes on correlation results. Lastly, we show the reduction in detection threshold and analyst workload offered by waveform correlation techniques compared to STA/LTA based detection. We analyzed 4 days of data from each aftershock sequence using all three methods. Most known events clustered in a similar manner across the toolsets. Up to 25% of catalogued events were found to be a member of a cluster. In addition, the Waveform Correlation Detector and Self Scanner identified significant numbers of new events that were not in either the EDR or regional catalogs, showing a lowering of the detection threshold. We extended our analysis to study the effect of distance on correlation results by applying the analysis tools to multiple stations along a transect of nearly constant azimuth when possible. We expected the number of events found via correlation would drop off as roughly 1/r2, where r is the distance from mainshock to station. However, we found that regional geological conditions influenced the performance of a given station more than distance. For example, for one sequence we clustered 25% of events at the nearest station to the mainshock (34 km), while our performance dropped to 2% at a station 550 km distant. However, we matched our best performance (25% clustering) at a station 198 km distant.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

Instrumental Backgrounds to $t\bar{t}$ and Single Top Production at Hadron Colliders

A number of backgrounds to top-quark production at the LHC and Tevatron experiments are characterized as instrumental: beam related-backgrounds, cosmic ray backgrounds, and backgrounds from non-prompt (or fake) leptons. A number of strategies have been deployed in the experiments to minimize these backgrounds and to estimate their rates. The strategies are applied at the level of the accelerator, the detector apparatus, the first few cuts which are common to many physics analyses, and more refined kinematic cuts. Remaining fake lepton background is estimated using sophisticated data-driven techniques. We present here a survey of the state-of-the-art in the reduction of instrumental backgrounds and also their estimation.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Fault Detection and Diagnosis In Hall–Héroult Cells Based on Individual Anode Current Measurements Using Dynamic Kernel PCA

Individual anode current signals in aluminum reduction cells provide localized cell conditions in the vicinity of each anode, which contain more information than the conventionally measured cell voltage and line current. One common use of this measurement is to identify process faults that can cause significant changes in the anode current signals. While this method is simple and direct, it ignores the interactions between anode currents and other important process variables. This paper presents an approach that applies multivariate statistical analysis techniques to individual anode currents and other process operating data, for the detection and diagnosis of local process abnormalities in aluminum reduction cells. Specifically, since the Hall–Héroult process is time-varying with its process variables dynamically and nonlinearly correlated, dynamic kernel principal component analysis with moving windows is used. The cell is discretized into a number of subsystems, with each subsystem representing one anode and cell conditions in its vicinity. The fault associated with each subsystem is identified based on multivariate statistical control charts. The results show that the proposed approach is able to not only effectively pinpoint the problematic areas in the cell, but also assess the effect of the fault on different parts of the cell.

36 MATERIALS SCIENCE↗

Sky Residual Correction

The near-infrared is populated by numerous emission lines radiated from the Earth’s atmosphere, and these emission lines are often several orders of magnitude brighter in intensity than the typical astrophysical science target. The subtraction of these emission lines, sky subtraction, can create large systematic errors in ground-based astronomical spectra effectively limiting the number of usable resolution elements. A more effective sky subtraction, and the reduction of the systematic errors due to the sky subtraction process, is a major hurdle which ground-based astronomy must overcome to increase the amount of usable spectrum, and to be able to observe fainter scientific targets. Large high-quality data sets such as the Sloan Digital Sky Survey (SDSS) in themselves present opportunities for a reduction of the systematic sky subtraction errors through self-calibration. The sky residual correction method of Wild and Hewett is one such self-calibration technique which uses principal component analysis to reduce the systematic sky residual errors present in SDSS spectra. We apply sky residual corrections to the SDSS, Baryon Oscillation Spectroscopic Survey, and Apache Point Observatory Galactic Evolution Experiment data sets to optimize the number of subtraction components, and to quantify the reduction of the systematic errors in the science spectra. Finally as a proof of concept we use the sky-residual-corrected SDSS luminous red galaxy spectra to search for gravitationally lensed emission line Galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

Review of non destructive assay techniques utilised for decommissioning of plutonium facilities at AWE - 14418

Over the past fifteen years AWE has decommissioned and size reduced a large number of glove boxes that contained a wide spectrum of equipment that ranged from highly radiation attenuating structures, such as furnaces and lathes, to other less radiation attenuating apparatus, such as small discrete tools or simple work benches. In order to generate meaningful glove box residual fissile material inventory data, which can constructively inform the selection of measures needed to underwrite the criticality safety of size reduction and consignment to waste activities, several different types of Non Destructive Assay (NDA) measurement techniques are available at AWE. These are deployed for the estimate of the residual fissile material inventory prior to decommissioning activities and for the final NDA of waste containers, post decommissioning. The techniques range from the Decommissioning In-Situ Plutonium Inventory Monitor (DISPIM), through other Neutron Coincidence Counting (NCC) techniques (Neutron Panel monitoring) and gamma measurements to consideration of Thermo Luminescent Detector (TLD) data. This paper will present a comparison of the fissile material identified in the waste generated post decommissioning operations to the NDA estimate of the residual fissile material inventory for the glove boxes, as well as provide a direct comparison between the different NDA techniques available at AWE. The findings to date indicate that the glove box residual fissile material inventory used for criticality safety purposes bounds the decommissioning process, subject to the suitable deployment of the NDA technique and the inclusion of appropriate uncertainties. (authors)

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Investigating nearby star-forming galaxies in the ultraviolet with HST/COS spectroscopy. I. Spectral analysis and interstellar abundance determinations

This is the first in a series of three papers describing a project with the Cosmic Origins Spectrograph on the Hubble Space Telescope to measure abundances of the neutral interstellar medium (ISM) in a sample of nine nearby star-forming galaxies. The goal is to assess the (in)homogeneities of the multiphase ISM in galaxies where the bulk of metals can be hidden in the neutral phase, yet the metallicity is inferred from the ionized gas in the H II regions. The sample, spanning a wide range in physical properties, is to date the best suited to investigate the metallicity behavior of the neutral gas at redshift z = 0. ISM absorption lines were detected against the far-ultraviolet spectra of the brightest star-forming region(s) within each galaxy. Here we report on the observations, data reduction, and analysis of these spectra. Column densities were measured by a multicomponent line-profile fitting technique, and neutral-gas abundances were obtained for a wide range of elements. Several caveats were considered, including line saturation, ionization corrections, and dust depletion. Ionization effects were quantified with ad hoc CLOUDY models reproducing the complex photoionization structure of the ionized and neutral gas surrounding the UV-bright sources. An 'average spectrum of a redshift z = 0 star-forming galaxy' was obtained from the average column densities of unsaturated profiles of neutral-gas species. This template can be used as a powerful tool for studies of the neutral ISM at both low and high redshift.

79 ASTRONOMY AND ASTROPHYSICS↗

Surrogate models for plasma displacement and current in 3D perturbed magnetohydrodynamic equilibria in tokamaks

Abstract A numerical database of over one thousand perturbed three-dimensional (3D) equilibria has been generated, constructed based on the MARS-F (Liu et al 2000 Phys. Plasmas 7 3681) computed plasma response to the externally applied 3D field sources in multiple tokamak devices. Perturbed 3D equilibria with the n = 1–4 ( n is the toroidal mode number) toroidal periodicity are computed. Surrogate models are created for the computed perturbed 3D equilibrium utilizing model order reduction (MOR) techniques. In particular, retaining the first few eigenstates from the singular value decomposition (SVD) of the data is found to produce reasonably accurate MOR-representations for the key perturbed quantities, such as the perturbed parallel plasma current density and the plasma radial displacement. SVD also helps to reveal the core versus edge plasma response to the applied 3D field. For the database covering the conventional aspect ratio devices, about 95% of data can be represented by the truncated SVD-series with inclusion of only the first five eigenstates, achieving a relative error (RE) below 20%. The MOR-data is further utilized to train neural networks (NNs) to enable fast reconstruction of perturbed 3D equilibria, based on the two-dimensional equilibrium input and the 3D source field. The best NN-training is achieved for the MOR-data obtained with a global SVD approach, where the full set of samples used for NN training and testing are stretched and form a large matrix which is then subject to SVD. The fully connected multi-layer perceptron, with one or two hidden layers, can be trained to predict the MOR-data with less than 10% RE. As a key insight, a better strategy is to train separate NNs for the plasma response fields with different toroidal mode numbers. It is also better to apply MOR and to subsequently train NNs separately for conventional and low aspect ratio devices, due to enhanced toroidal coupling of Fourier spectra in the plasma response in the latter case.

3D equilibrium↗