Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “network analysis”

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 1,459 records · Page 81
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Data and Scripts associated with “Lambda-PFLOTRAN: Workflow for Incorporating Organic Matter Chemistry Informed by Ultra High Resolution Mass Spectrometry into Biogeochemical Modeling.”

This data package is associated with the publication “Lambda-PFLOTRAN: Workflow for Incorporating Organic Matter Chemistry Informed by Ultra High Resolution Mass Spectrometry into Biogeochemical Modeling” submitted to Geoscientific Model Development (Muller et al., 2024). In this manuscript, organic matter chemistry and thermodynamics are directly connected to reactive transport simulators through the newly developed Lambda-PFLOTRAN (Parallel Reactive Flow and Transport model) workflow tool that succinctly incorporates organic matter chemistry data generated from Fourier transform ion cyclotron resonance mass spectrometry (FTICR-MS) into reaction networks to simulate aerobic respiration of the organic matter and the resulting biogeochemistry. Lambda-PFLOTRAN is a python-based workflow, executed through a Jupyter Notebook interface, that digests raw FTICR-MS data, develops a representative reaction network based on substrate-explicit thermodynamic modeling (also termed lambda modeling due to its key thermodynamic parameter λ used therein), and completes a biogeochemical simulation with the open source, reactive flow, and transport code PFLOTRAN. This data package contains Jupyter Notebook based workflows for two test cases for running biogeochemical simulations of organic matter oxidation identified by FTICR-MS. It contains four primary folders (workflow, data, src, and analysis), a file-level metadata file (Muller_2024_Lambda_PFLOTRAN_Manuscript_Data_Package_flmd.csv) that lists all the files contained in this data package with a short description of each, and a data dictionary (Muller_2024_Lambda_PFLOTRAN_Manuscript_Data_Package_dd.csv) file that describes the tabular column headers. The ‘workflow’ folder contains the Jupyter Notebook based workflows for running the lambda analysis, PFLOTRAN simulation, sensitivity analysis and parameter estimation. The ‘data’ folder contains the FTICR-MS data, initial conditions, and incubation data for test cases 1 and 2 in folders titled ‘WHONDRS’ and ‘Colloids’, respectively. The data folder also has a ‘Database’ folder containing a reaction network for bulk organic matter (assumed to be CH2O) and a general database for PFLOTRAN (hanford_rxn_network). The CH2O reaction network defines bulk organic matter oxidation. Biogeochemical simulations are completed for both the lambda binned organic matter and bulk organic matter reaction networks. The ‘hanford_rxn_network’ database includes information required for PFLTORAN simulations including ion size, molar mass, and charge of the aqueous species, gases, and minerals phases. The ‘src’ folder contains python source codes for performing lambda analysis, PFLOTRAN simulation, sensitivity analysis and parameter estimation. The ‘analysis’ folder contains outputs from the test cases 1 and 2 including lambda analysis, PFLOTRAN runs and the calibration results.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Solving differential equations using deep neural networks

Recent work on solving partial differential equations (PDEs) with deep neural networks (DNNs) is presented. The paper reviews and extends some of these methods while carefully analyzing a fundamental feature in numerical PDEs and nonlinear analysis: irregular solutions. First, the Sod shock tube solution to the compressible Euler equations is discussed and analyzed. This analysis includes a comparison of a DNN-based approach with conventional finite element and finite volume methods, and demonstrates that the DNN is competitive in terms of degrees of freedom required for a given accuracy. Further, the DNN-based approach is extended to consider performance improvements and simultaneous parameter space exploration. Next, a shock solution to compressible magnetohydrodynamics (MHD) is solved for, and used in a scenario where experimental data is utilized to enhance a PDE system that is a priori insufficient to validate against the observed/experimental data. This is accomplished by enriching the model PDE system with source terms that are then inferred via supervised training with synthetic experimental data. The resulting DNN framework for PDEs enables straightforward system prototyping and natural integration of large data sets (be they synthetic or experimental), all while simultaneously enabling single-pass exploration of an entire parameter space.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2021

Transport Upscaling under Flow Heterogeneity and Matrix-Diffusion in Three-Dimensional Discrete Fracture Networks

For this work, we investigate the combined effects of network scale flow variability and retention due to matrix-diffusion on the scaling behavior of transport through fractured media. Two of the principal mechanisms controlling the transport of solutes through fractured low-permeability media are broad distributions of flow velocities and retention times in the solid matrix. We study the relative impact of these two processes under different initial conditions using a set of three-dimensional discrete fracture network simulations. We use these simulations to develop and calibrate an upscaled continuous time random walk (CTRW) approach for advective transport based on an Ornstein-Uhlenbeck model for the particle velocities that accounts for the fracture-matrix coupling using a compound Poisson process. This CTRW model can be conditioned on the initial solute distribution and allows to observe late-time scaling behavior at distances beyond what is feasible to observe using high-fidelity direct numerical simulations. We determine that the initial distribution of particles leads to marked differences in the persistent long-term scale behavior in the solute travel time distributions, even those undergoing retention due to matrix diffusion through implementation and analysis of the model.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Deep learning to estimate permeability using geophysical data

Time-lapse electrical resistivity tomography (ERT) is a popular geophysical method to estimate three-dimensional (3D) permeability fields from electrical potential difference measurements. Traditional inversion and data assimilation methods are used to ingest this ERT data into hydrogeophysical models to estimate permeability. Due to ill-posedness and the curse of dimensionality, existing inversion strategies provide poor estimates and low resolution of the 3D permeability field. Recent advances in deep learning provide us with powerful algorithms to overcome this challenge. This paper presents a deep learning (DL) framework to estimate the 3D subsurface permeability from time-lapse ERT data. To test the feasibility of the proposed framework, we train DL-enabled inverse models on simulation data. Each measurement in both synthetic and field data is standardized by removing the mean and scaling the time-series to unit variance. This pre-processing step is necessary to bring simulation data closer to field observations. Subsurface process models based on hydrogeophysics are used to generate this synthetic data. Training performed on limited simulation data resulted in the DL model over-fitting. An advanced data augmentation based on mixup is implemented to generate additional training samples to overcome this issue. This mixup technique creates weakly labeled (low-fidelity) samples from strongly labeled (high-fidelity) data. The weakly labeled training data is then used to develop DL-enabled inverse models and reduce over-fitting. As both time-lapse ERT (1133048 features/realization) and 3D permeability (585453 features/realization) data samples are from a high-dimensional space, principal component analysis (PCA) is employed to reduce dimensionality. Encoded ERT and encoded permeability are generated using the trained PCA estimators. A deep neural network is then trained to map the encoded ERT to encoded permeability. This mixup training and unsupervised learning allowed us to build a fast and reasonably accurate DL-based inverse model under limited simulation data. Results show that proposed weak supervised learning can capture salient spatial features in the 3D permeability field. Quantitatively, the average mean squared error (in terms of the natural log) on the strongly labeled training, validation, and test datasets is less than 0.5. The R 2 -score (global metric) is greater than 0.75, and the percent error in each cell (local metric) is less than 10%. Finally, an added benefit in terms of computational cost is that the proposed DL-based inverse model is at least O(10 4 ) times faster than running a forward model once it is trained. Data generation, DL model training, and hyperparameter tuning to identify optimal neural network architectures utilized high-performance computing resources while the DL inference is performed on a standard laptop. Approximately, O(10 5 ) processor hours are used for generating data and DL tuning and training. We acknowledge that the data generation and DL model development are expensive. But once a DL model is trained, it can be re-used for inversion rapidly for the given system, with set physics and domain. Note that traditional inversion may require multiple forward model simulations (e.g., in the order of 10 to 1000), which are very expensive. This computational savings ≈ O(10 5 ) – O(10 7 )) makes the proposed DL-based inverse model attractive for subsurface imaging and real-time ERT monitoring applications due to fast and yet reasonably accurate estimations of permeability field.

58 GEOSCIENCES↗
DOE OSTI2021

Adoption of perennial energy crops in the US Midwest: Causal and heterogeneous determinants

The success of a bioenergy policy relies largely on the wide adoption of perennial energy crops at the farm-scale. This study uses survey data to examine potential adoption decisions by farmers in the US Midwest and the causal effects of various direct and indirect influencing factors, especially heterogeneous preferences of farmers. A Bayesian network (BN) model is developed to delineate the causal-relationship between farmers adoption decisions and the influencing factors. Here, we find a dominating role of economic factors and a non-negligible impact of non-economic factors, such as the perceived environmental benefits and the extent of familiarity with perennial energy crops. To examine the effect of heterogeneity in farmer preferences, we classify the surveyed farmers into four categories based on their attitudes toward the economic, social, and environmental dimensions of perennial energy crops. We identify statistically significant between-group differences in the responses of the four types of farmers to the various influencing factors. Our findings contribute to disentangling the complicated motivations that will influence perennial energy crop adoption decisions and provide implications for more targeted policy development that need to consider the heterogeneous drivers of farmer decisions about land use.

09 BIOMASS FUELS↗
DOE OSTI2025

Parallel sorting algorithm classification: is manual instrumentation necessary?

Understanding parallel algorithms is crucial for accelerating scientific simulations on complex, distributed memory, high-performance computers. Modern algorithm classification approaches learn semantics directly from source code to differentiate between algorithms, however, accessing source code is not always possible. We can learn about parallel algorithms from observing their performance, as programs running the same algorithms and using the same hardware should exhibit similar performance characteristics. We present an approach to learn algorithm classes from parallel performance data directly in order to classify algorithms without access to the source code. We extend previous work to enable classifying parallel sorting algorithms using automatic instrumentation instead of requiring manual region annotations in the source code. In this work, we design and demonstrate a study for classification of parallel sorting algorithms using parallel performance data collected from automatic instrumentation, and evaluate the performance of our new methodology on classification. We leverage Caliper to collect the performance data, Thicket for our exploratory data analysis (EDA), and PyTorch and Scikit-learn to evaluate the effectiveness of random forests, support vector machines (SVMs), decision trees, neural networks, and logistic regressions on parallel performance data. Additionally, we study noise in parallel performance data, whether the removal of noise and pre-processing of the data is necessary to accurately classify parallel sorting algorithms, and determine the effectiveness of features created from performance data. In conclusion, we demonstrate classification accuracy for these five different models of up to 97.7% across four different parallel algorithm classes.

Algorithm Classification↗
DOE OSTI2021

Application of remote sensing precipitation data and the $\mathrm{CONNECT}$ algorithm to investigate spatiotemporal variations of heavy precipitation: Case study of major floods across Iran (Spring 2019)

In recent years, the number of floods following unprecedented rainfall events have increased in Iran during early spring (March 21st to April 20th, referred to in Iran as the month of “Farvadin”). While numerous studies have addressed changes in climate extremes and precipitation trends at different temporal scales from daily to annual across the country, analyses of short-duration and heavy precipitation, especially during recent years, are rarely considered. Furthermore, most studies investigate the variations in extremes and total precipitation using a limited number of synoptic weather stations across Iran. Here this study assesses the variations in heavy precipitation (precipitation with intensities greater than or equal to 3 mm/3 h) at 0.04° spatial and 3-hourly temporal resolution during the month of Farvardin. In addition, the effect of atmospheric river conditions over Iran and their possible link to intensifying heavy precipitation is explored. For this purpose, the CONNected-objECT (CONNECT) algorithm is applied on a precipitation dataset, Precipitation Estimation from Remotely Sensed Information Using Artificial Neural Networks-Cloud Classification System (PERSIANN-CCS), and an Integrated Water Vapor Transport (IVT) dataset from the NASA Modern-Era Retrospective Analysis for Research and Applications Version-2 (MERRA-2). The results suggest that the increase in the number of floods in recent years is related to the increase in the intensity and volume of heavy precipitation events, although the frequency and duration of heavy precipitation events have not changed significantly. Furthermore, the results show that atmospheric river conditions over the country are present during the same window as each year’s most extreme events. It is found that 8 out of 13 of the largest ARs over Iran come from moisture plumes with pathways over the African and Red Sea.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Machine learning-based real-time monitoring system for smart connected worker to improve energy efficiency

Recent advances in machine learning and computer vision brought to light technologies and algorithms that serve as new opportunities for creating intelligent and efficient manufacturing systems. In this study, the real-time monitoring system of manufacturing workflow for the Smart Connected Worker (SCW) is developed for the small and medium-sized manufacturers (SMMs), which integrates state-of-the-art machine learning techniques with the workplace scenarios of advanced manufacturing systems. Specifically, object detection and text recognition models are investigated and adopted to ameliorate the labor-intensive machine state monitoring process, while artificial neural networks are introduced to enable real-time energy disaggregation for further optimization. The developed system achieved efficient supervision and accurate information analysis in real-time for prolonged working conditions, which could effectively reduce the cost related to human labor, as well as provide an affordable solution for SMMs. The competent experiment results also demonstrated the feasibility and effectiveness of integrating machine learning technologies into the realm of advanced manufacturing systems.

42 ENGINEERING↗
DOE OSTI2025

Three-Dimensional Pore Networks in Miocene Stevens Sandstone of California: Implications for CO 2 Geologic Storage

The Miocene Stevens Sandstone in the San Joaquin Basin of California is increasingly recognized as a promising candidate for CO 2 geological storage due to the enormous storage capacity, proven sealing, and existing infrastructure. In this study, computed microtomography imaging and pore network modeling were employed to investigate the influence of pore geometry and wettability on the CO 2 injectivity and residual trapping. Image analysis revealed that a significant fraction of the cement and matrix consists of microporous regions. The microporosity can substantially increase the overall pore space, yet its contribution to permeability remains modest, particularly in samples with low permeability. The intrinsic heterogeneity of turbidite reservoirs further complicates the reservoir properties among different layers. Two-phase flow simulations under varying wettability conditions (water-wet, weak water-wet, and neutral-wet) demonstrated that the CO 2 injection is predominantly controlled by macropores. CO 2 invades microporous regions only after these larger pores are filled. The presence of microporosity leads to a decrease in both initial and residual CO 2 saturations, with the magnitude of the reduction being influenced by wettability. Neutral-wet scenarios exhibit higher CO 2 mobility and thus lower residual trapping than water-wet scenarios. The results imply that heterogeneity in pore geometry and cement distribution across different layers can result in stratified CO 2 flow pathways, complicating efforts to predict injection performance. Overall, the Stevens Sandstone shows considerable promise for CO 2 geologic storage, but effective implementation will require detailed characterization of the pore structure as well as the integration of reactive fluid flow to account for potential mineral dissolution and fines migration.

fluids↗
DOE OSTI2026

Sequential Fracture Activation and Stress Evolution During EGS Stimulation at Utah FORGE Revealed by Waveform Cross‐Correlation

Mapping fracture networks in Enhanced Geothermal Systems (EGS) is essential for optimizing reservoir performance, yet complex fracture evolution during stimulation remains difficult to resolve. This study examines the evolution of microseismicity and fracture networks during stage 3 of the 2022 EGS stimulation at the Utah Frontier Observatory for Research in Geothermal Energy site. We map the fracture network represented by 20 clusters of seismic events identified by waveform similarities with cross-correlation. We characterize their geometric properties such as strike, dip, length, and width, and analyze the time evolution of activated fractures. The results reveal a systematic fracture evolution: early activation of pre-existing natural fractures, complex network development during peak injection, and continued activation of less favorably oriented fractures post-injection. Magnitude calibration using the Principal Component Analysis of cross-correlated waveforms improves relative amplitude measurements, refining estimations of the Gutenberg-Richter b-values with spatial variations in b-values suggesting stress re-distribution across the stimulated area. Analysis of the stress state of selected fractures further shows that fractures requiring higher excess pore pressure primarily activate at the end of injection and post-injection, highlighting stress transfer due to pore pressure as a dominant triggering mechanism. These findings provide insights into fracture propagation, stress evolution, and seismic hazard assessment in EGS reservoirs.

Asirifi, Richard [Texas A & M Univ., College Stati↗
DOE OSTI2022

Adsorbate chemical environment-based machine learning framework for heterogeneous catalysis

Abstract Heterogeneous catalytic reactions are influenced by a subtle interplay of atomic-scale factors, ranging from the catalysts’ local morphology to the presence of high adsorbate coverages. Describing such phenomena via computational models requires generation and analysis of a large space of atomic configurations. To address this challenge, we present Adsorbate Chemical Environment-based Graph Convolution Neural Network (ACE-GCN), a screening workflow that accounts for atomistic configurations comprising diverse adsorbates, binding locations, coordination environments, and substrate morphologies. Using this workflow, we develop catalyst surface models for two illustrative systems: (i) NO adsorbed on a Pt 3 Sn(111) alloy surface, of interest for nitrate electroreduction processes, where high adsorbate coverages combined with low symmetry of the alloy substrate produce a large configurational space, and (ii) OH* adsorbed on a stepped Pt(221) facet, of relevance to the Oxygen Reduction Reaction, where configurational complexity results from the presence of irregular crystal surfaces, high adsorbate coverages, and directionally-dependent adsorbate-adsorbate interactions. In both cases, the ACE-GCN model, trained on a fraction (~10%) of the total DFT-relaxed configurations, successfully describes trends in the relative stabilities of unrelaxed atomic configurations sampled from a large configurational space. This approach is expected to accelerate development of rigorous descriptions of catalyst surfaces under in-situ conditions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Computational scanning tunneling microscope image database

We introduce the systematic database of scanning tunneling microscope (STM) images obtained using density functional theory (DFT) for two-dimensional (2D) materials, calculated using the Tersoff-Hamann method. It currently contains data for 716 exfoliable 2D materials. Examples of the five possible Bravais lattice types for 2D materials and their Fourier-transforms are discussed. All the computational STM images generated in this work are made available on the JARVIS-STM website ( https://jarvis.nist.gov/jarvisstm ). We find excellent qualitative agreement between the computational and experimental STM images for selected materials. As a first example application of this database, we train a convolution neural network model to identify the Bravais lattice from the STM images. We believe the model can aid high-throughput experimental data analysis. These computational STM images can directly aid the identification of phases, analyzing defects and lattice-distortions in experimental STM images, as well as be incorporated in the autonomous experiment workflows.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Constraints on primordial non-Gaussianity from the cross-correlation of DESI luminous red galaxies and Planck CMB lensing

We use the angular cross-correlation between a luminous red galaxy (LRG) sample from the Dark Energy Spectroscopic Instrument (DESI) Legacy Survey data release DR9 and the Planck cosmic microwave background (CMB) lensing maps to constrain the local primordial non-Gaussianity parameter, f NL , using the scale-dependent galaxy bias effect. The galaxy sample covers approximately 40% of the sky, contains galaxies up to redshift z ∼ 1.4, and is calibrated with the LRG spectra that have been observed for DESI Year 1 (Y1). We apply a nonlinear imaging systematics treatment based on neural networks to remove observational effects that could potentially bias the f NL measurement. Our measurement is performed without blinding, but the full analysis pipeline is tested with simulations including systematics. Using the two-point angular cross-correlation between LRG and CMB lensing only, we find f NL = 39 −38 +40 at the 68% confidence level, and our result is robust in terms of systematics and cosmological assumptions. If we combine this information with the autocorrelation of LRG, applying a scale cut to limit the impact of systematics, we find f NL = 24 −21 +20 at the 68% confidence level. Our results motivate the use of CMB lensing cross-correlations to measure f NL with future datasets, given its stability in terms of observational systematics compared to the angular autocorrelation. Furthermore, performing accurate systematics mitigation is crucially important in order to achieve competitive constraints on f NL from CMB lensing cross-correlation in combination with the tracers’ autocorrelation.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

EFIT-Prime: Probabilistic and physics-constrained reduced-order neural network model for equilibrium reconstruction in DIII-D

We introduce EFIT-Prime, a novel machine learning surrogate model for EFIT (Equilibrium FIT) that integrates probabilistic and physics-informed methodologies to overcome typical limitations associated with deterministic and ad hoc neural network architectures. EFIT-Prime utilizes a neural architecture search-based deep ensemble for robust uncertainty quantification, providing scalable and efficient neural architectures that comprehensively quantify both data and model uncertainties. Physically informed by the Grad–Shafranov equation, EFIT-Prime applies a constraint on the current density J tor and a smoothness constraint on the first derivative of the poloidal flux, ensuring physically plausible solutions. Furthermore, the spatial location of the diagnostics is explicitly incorporated in the inputs to account for their spatial correlation. Extensive evaluations demonstrate EFIT-Prime's accuracy and robustness across diverse scenarios, most notably showing good generalization on negative-triangularity discharges that were excluded from training. Timing studies indicate an ensemble inference time of 15 ms for predicting a new equilibrium, offering the possibility of plasma control in real-time, if the model is optimized for speed.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

An isolated water droplet in the aqueous solution of a supramolecular tetrahedral cage

Water under nanoconfinement at ambient conditions has exhibited low-dimensional ice formation and liquid–solid phase transitions, but with structural and dynamical signatures that map onto known regions of water’s phase diagram. Using terahertz (THz) absorption spectroscopy and ab initio molecular dynamics, we have investigated the ambient water confined in a supramolecular tetrahedral assembly, and determined that a dynamically distinct network of 9 ± 1 water molecules is present within the nanocavity of the host. The low-frequency absorption spectrum and theoretical analysis of the water in the Ga 4 L 6 12– host demonstrate that the structure and dynamics of the encapsulated droplet is distinct from any known phase of water. A further inference is that the release of the highly unusual encapsulated water droplet creates a strong thermodynamic driver for the high-affinity binding of guests in aqueous solution for the Ga 4 L 6 12– supramolecular construct.

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

Physics-guided machine learning approaches to predict the ideal stability properties of fusion plasmas

One of the biggest challenges to achieve the goal of producing fusion energy in tokamak devices is the necessity of avoiding disruptions of the plasma current due to instabilities. The disruption event characterization and forecasting (DECAF) framework has been developed in this purpose, integrating physics models of many causal events that can lead to a disruption. Two different machine learning approaches are proposed to improve the ideal magnetohydrodynamic (MHD) no-wall limit component of the kinetic stability model included in DECAF. First, a random forest regressor (RFR), was adopted to reproduce the DCON computed change in plasma potential energy without wall effects for a large database of equilibria from the national spherical torus experiment (NSTX). This tree-based method provides an analysis of the importance of each input feature, giving an insight into the underlying physics phenomena. Secondly, a fully-connected neural network has been trained on sets of calculations with the DCON code, to get an improved closed form equation of the no-wall β limit as a function of the relevant plasma parameters indicated by the RFR. The neural network has been guided by physics theory of ideal MHD in its extension outside the domain of the NSTX experimental data. The estimated value has been incorporated into the DECAF kinetic stability model and tested against a set of experimentally stable and unstable discharges. Moreover, the neural network results were used to simulate a real-time stability assessment using only quantities available in real-time. Finally, the portability of the model was investigated, showing encouraging results by testing the NSTX-trained algorithm on the mega ampere spherical tokamak (MAST).

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2020

Reionization history constraints from neural network based predictions of high-redshift quasar continua

ABSTRACT Observations of the early Universe suggest that reionization was complete by z ∼ 6, however, the exact history of this process is still unknown. One method for measuring the evolution of the neutral fraction throughout this epoch is via observing the Lyα damping wings of high-redshift quasars. In order to constrain the neutral fraction from quasar observations, one needs an accurate model of the quasar spectrum around Lyα, after the spectrum has been processed by its host galaxy but before it is altered by absorption and damping in the intervening intergalactic medium (IGM). In this paper, we present a novel machine learning approach, using artificial neural networks, to reconstruct quasar continua around Lyα. Our Quasar Spectra from Artificial Neural Network based predictive Regression Algorithm(QSANNdRA) improves the error in this reconstruction compared to the state-of-the-art principal component analysis (PCA) based model in the literature by 14.2 per cent on average, and provides an improvement of 6.1 per cent on average when compared to an extension thereof. In comparison with the extended PCA model, QSANNdRA further achieves an improvement of 22.1 per cent and 16.8 per cent when evaluated on low-redshift quasars most similar to the two high-redshift quasars under consideration, ULAS J1120+0641 at z = 7.0851 and ULAS J1342+0928 at z = 7.5413, respectively. Using our more accurate reconstructions of these two z > 7 quasars, we estimate the neutral fraction of the IGM using a homogeneous reionization model and find $\bar{x}_\mathrm{H\, \small{I}} = 0.25^{+0.05}_{-0.05}$ at z = 7.0851 and $\bar{x}_\mathrm{H\, \small{I}} = 0.60^{+0.11}_{-0.11}$ at z = 7.5413. Our results are consistent with the literature and favour a rapid end to reionization.

Ďurovčíková, Dominika↗