Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “datasets”

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 1,639 records · Page 91
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Adiabatic quantum support vector machines

Adiabatic quantum computers can solve difficult optimization problems (e.g., the quadratic unconstrained binary optimization problem), and they seem well suited to train machine learning models. In this paper, we describe an adiabatic quantum approach for training support vector machines. We show that the time complexity of our quantum approach is an order of magnitude better than the classical approach. Next, we compare the test accuracy of our quantum approach against a classical approach that uses the Scikit-learn library in Python across five benchmark datasets (Iris, Wisconsin Breast Cancer (WBC), Wine, Digits, and Lambeq). We show that our quantum approach obtains accuracies on par with the classical approach. Finally, we perform a scalability study in which we compute the total training times of the quantum approach and the classical approach with an increasing number of features and an increasing number of data points in the training dataset. In conclusion, our scalability results show that the quantum approach obtains a 3.5–4.5x speedup over the classical approach on datasets with many (millions of) features.

Computational Complexity↗
DOE OSTI2023

DownScaleBench for developing and applying a deep learning based urban climate downscaling- first results for high-resolution urban precipitation climatology over Austin, Texas

Abstract Cities need climate information to develop resilient infrastructure and for adaptation decisions. The information desired is at the order of magnitudes finer scales relative to what is typically available from climate analysis and future projections. Urban downscaling refers to developing such climate information at the city (order of 1 – 10 km) and neighborhood (order of 0.1 – 1 km) resolutions from coarser climate products. Developing these higher resolution (finer grid spacing) data needed for assessments typically covering multiyear climatology of past data and future projections is complex and computationally expensive for traditional physics-based dynamical models. In this study, we develop and adopt a novel approach for urban downscaling by generating a general-purpose operator using deep learning. This ‘DownScaleBench’ tool can aid the process of downscaling to any location. The DownScaleBench has been generalized for both in situ (ground- based) and satellite or reanalysis gridded data. The algorithm employs an iterative super-resolution convolutional neural network (Iterative SRCNN) over the city. We apply this for the development of a high-resolution gridded precipitation product (300 m) from a relatively coarse (10 km) satellite-based product (JAXA GsMAP). The high-resolution gridded precipitation datasets is compared against insitu observations for past heavy rain events over Austin, Texas, and shows marked improvement relative to the coarser datasets relative to cubic interpolation as a baseline. The creation of this Downscaling Bench has implications for generating high-resolution gridded urban meteorological datasets and aiding the planning process for climate-ready cities.

Singh, Manmeet (ORCID:0000000233747149)↗
DOE OSTI2022

What you get is not always what you see—pitfalls in solar array assessment using overhead imagery

Effective integration planning for small, distributed solar photovoltaic (PV) arrays into electric power grids requires access to high quality data: the location and power capacity of individual solar PV arrays. Unfortunately, national databases of small-scale solar PV do not exist; those that do are limited in their spatial resolution, typically aggregated up to state or national levels. While several promising approaches for solar PV detection have been published, strategies for evaluating the performance of these models are often highly heterogeneous from study to study. The resulting comparison of these methods for practical applications for energy assessments becomes challenging and may imply that the reported performance evaluations overly optimistic. The heterogeneity comes in many forms, each of which we explore in this work: the degree of diversity of the locations and sensors (e.g. different satellites, aerial photography) from which the training and validation data originate, the validation of ground truth (manual annotation of imagery vs known solar PV locations), the level of spatial aggregation (e.g. array-level vs regional estimates), and inconsistencies in the training and validation datasets (e.g. different datasets are used for each study and those data are not always made accessible). For each, we discuss emerging practices from the literature to address them or suggest directions of future research. As part of our investigation, we evaluate solar PV identification performance in two large regions: the entire state of Connecticut and the city of San Diego, CA. In Connecticut, we also use 33,114 known parcel-level solar PV installations from Berkeley Lab’s Tracking the Sun dataset to evaluate parcel-level performance and evaluate capacity estimates using 169 municipalities. We also make our code (which we call SolarMapper), pre-trained models, training data, and predictions publicly available and provide a web portal for interactively inspecting each prediction that was made. Here our findings suggest that traditional performance evaluation of the automated identification of solar PV from satellite imagery may be optimistic due to common limitations in the validation process. The takeaways from this work are intended to inform and catalyze the large-scale practical application of automated solar PV assessment techniques by energy researchers and professionals.

14 SOLAR ENERGY↗
DOE OSTI2024

Development of typical solar years and typical wind years for efficient assessment of renewable energy systems across the U.S.

Weather data plays a critical role in renewable energy analysis. Compared to using multiple Actual Meteorological Years, simulations using a single typical year require significantly fewer computational resources. Previous efforts to create typical weather datasets for renewable energy analysis either lack justified or optimized strategies for selecting and weighting different weather parameters or are limited to a few specific locations. Here, in this study, we developed a dataset comprising Typical Solar Years (TSYs) and Typical Wind Years (TWYs) for over 2000 locations across the U.S., based on data from NASA's POWER project. The strategies for creating TSYs and TWYs were optimized based on the simulated outputs of various PV systems and wind turbines in 16 representative cities. This dataset provides an efficient means for the rapid evaluation and optimization of renewable energy systems throughout the entire U.S. Additionally, the optimal strategies identified in this study can be directly applied to create near-optimal TSYs and TWYs for most locations worldwide. However, readers can also employ the optimization approach presented in this work to develop optimal strategies tailored for particular regions.

NASA POWER↗
DOE OSTI2021

Industrial control system device classification using network traffic features and neural network embeddings

Characterization of modern cyber–physical Industrial Control System (ICS) devices is critical to the evaluation of their security posture and an understanding of the underlying industrial processes with which they interact. In this work, we address two related ICS device identification tasks: (1) separating ICS from non-ICS devices and (2) identifying specific ICS device types. We propose two distinct methods (one based on the existing IP2Vec method, and a novel traffic-features-based method) for achieving the first task. For transferability of the first task between two datasets, the traffic-features-based method performs significantly better (75% overall accuracy) compared to IP2Vec (22.5% overall accuracy). We further propose a novel method called DNP2Vec to address the second task. DNP2Vec is evaluated on two different datasets and achieves perfect multi-class classification accuracy (100%) for both datasets.

42 ENGINEERING↗
DOE OSTI2019

Building fault detection and diagnostics: Achieved savings, and methods to evaluate algorithm performance

Fault detection and diagnosis (FDD) represents one of the most active areas of research and commercial product development in the buildings industry. This paper addresses two questions concerning FDD implementation and advancement 1) What are today's users of FDD saving and spending on the technology? 2) What methods and datasets can be used to evaluate and benchmark FDD algorithm performance? Relevant to the first question, 26 organizations that use FDD across a total 550 buildings and 97 M sf achieved median savings of 8%. Twenty-seven FDD users reported that the median base cost for FDD software, annual recurring software cost, and annual labor cost were $8, $2.7 and $8 per monitoring point, with a median implementation size of approximately 1300 points. To address the second question, this paper describes a systematic methodology for evaluating the performance of FDD algorithms, curates an initial test dataset of air handling unit (AHU) system faults, and completes a trial to demonstrate the evaluation process on three sample FDD algorithms. The work provided a first step toward a standard evaluation of different FDD technologies. Finally, it showed the test methodology is indeed scalable and repeatable, provided an understanding of the types of insights that can be gained from algorithm performance testing, and highlighted the priorities for further expanding the test dataset.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2022

BASIN-3D: A brokering framework to integrate diverse environmental data

Diverse observational and simulation datasets are needed to understand and predict complex ecosystem behavior over seasonal to decadal and century time-scales. Integration of these datasets poses a major barrier towards advancing environmental science, particularly due to differences in the structure and formats of data provided by various sources. Here, we describe BASIN-3D (Broker for Assimilation, Synthesis and Integration of eNvironmental Diverse, Distributed Datasets), a data integration framework designed to dynamically retrieve and transform heterogeneous data from different sources into a common format to provide an integrated view. BASIN-3D enables users to adopt a standardized approach for data retrieval and avoid customizations for the data type or source. We demonstrate the value of BASIN-3D with two use cases that require integration of data from regional to watershed spatial scales. The first application uses the BASIN-3D Python library to integrate time-series hydrological and meteorological data to provide standardized inputs to analytical and machine learning codes in order to predict the impacts of hydrological disturbances on large river corridors of the United States. The second application uses the BASIN-3D Django framework to integrate diverse time-series data in a mountainous watershed in East River, Colorado, United States to enable scientific researchers to explore and download data through an interactive web portal. Thus, BASIN-3D can be used to support data integration for both web-based tools, as well as data analytics using Python scripting and extensions like Jupyter notebooks. The framework is expected to be transferable to and useful for many other field and modeling studies.

Varadharajan, C↗
DOE OSTI2022

Prediction of local concentration fields in porous media with chemical reaction using a multi scale convolutional neural network

The study of solute transport in porous media is of interest in many chemical engineering systems. Some example applications include packed bed catalytic reactors, filtration devices, and batteries. The pore scale modeling of these systems is time consuming and may require large computing resources, for this reason computational fluid dynamics (CFD) simulations are not practical if a large number of simulations is required, like in multiscale modeling, where a model at a large scale calls for pore scale simulations. It has been shown that neural networks can be trained with a dataset of flow simulations and then predict fields orders of magnitude faster, and with less computational resources, in new domains. However, it is crucial to provide the neural network with an effective description of the domain and the undergoing operating conditions to be able to train models that generalize accurately in unseen samples. Therefore, research is needed to employ neural networks in new complex systems. The appropriate training of a network for predicting coupled flow and solute transport processes is an outstanding problem due to the complex interplay between geometry and operating conditions. In this work, we train a multi scale convolutional neural network (MSNet) with a diverse dataset of simulations of transport and chemical reaction in porous media to predict the local concentration fields in images of porous media. Our dataset contains a wide diversity of sphere pack arrangements under different operating conditions (Péclet and Reynolds numbers). Further, we train a robust model by employing different input descriptors that represent the medium and the different operating conditions of each system. Our trained model is able to provide nearly instantaneous predictions, compared to around twenty hours of the CFD workflow, with less than 3.5% error on new geometries and transport conditions. Thus the model could be easily integrated in a multiscale workflow where fast response is needed.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Transient anisotropic kernel for probabilistic learning on manifolds

PLoM (Probabilistic Learning on Manifolds) is a method introduced in 2016 for handling small training datasets by projecting an Itô equation from a stochastic dissipative Hamiltonian dynamical system, acting as the MCMC generator, for which the KDE-estimated probability measure with the training dataset is the invariant measure. PLoM performs a projection on a reduced-order vector basis related to the training dataset, using the diffusion maps (DMAPS) basis constructed with a time-independent isotropic kernel. In this paper, we propose a new ISDE projection vector basis built from a transient anisotropic kernel, providing an alternative to the DMAPS basis to improve statistical surrogates for stochastic manifolds with heterogeneous data. The construction ensures that for times near the initial time, the DMAPS basis coincides with the transient basis. For larger times, the differences between the two bases are characterized by the angle of their spanned vector subspaces. The optimal instant yielding the optimal transient basis is determined using an estimation of mutual information from Information Theory, which is normalized by the entropy estimation to account for the effects of the number of realizations used in the estimations. Consequently, this new vector basis better represents statistical dependencies in the learned probability measure for any dimension. Three applications with varying levels of statistical complexity and data heterogeneity validate the proposed theory, showing that the transient anisotropic kernel improves the learned probability measure.

Diffusion maps↗
DOE OSTI2025

From clutter to clarity: Emergent neural operators via questionnaire metrics

Real-world datasets in chemical engineering and bioengineering processes—such as those from catalytic reactors, multiphase flows, polymerization reactors, bioreactors, and clinical trials—can often be unlabeled or disorganized, rendering the training of existing supervised learning models ineffective at learning the underlying dynamics. To salvage these datasets for decision-making, we first seek to obtain clarity from the cluttered data. Here, we present a framework for developing “structural” generative models, discovering emergent equations, and constructing efficient emulators from scrambled datasets by integrating unsupervised organizational learning techniques (Questionnaires) with advanced deep learning architectures (Deep Hidden Physics Models and Deep Operator Networks). Our approach is demonstrated on two illustrative model systems: (a) a 1D advection–diffusion partial differential equation representing a winding underground pipe and (b) an ensemble of Stuart–Landau oscillators, an agent-based system of coupled ordinary differential equations. In both cases, we successfully reconstruct meaningful spatial, temporal, and parameter embeddings from scrambled data, enabling good predictions of system dynamics. As a result, we highlight the framework’s potential for broader applications, enabling data-driven system identification in fields with inherently disorganized or hidden parameter spaces.

42 ENGINEERING↗
DOE OSTI2025

A network-enabled pipeline for gene discovery and validation in non-model plant species

Identifying key regulators of important genes in non-model crop species is challenging due to limited multi-omics resources. To address this, we introduce the network-enabled gene discovery pipeline NEEDLE, a user-friendly tool that systematically generates coexpression gene network modules, measures gene connectivity, and establishes network hierarchy to pinpoint key transcriptional regulators from dynamic transcriptome datasets. After validating its accuracy with two independent datasets, we applied NEEDLE to identify transcription factors (TFs) regulating the expression of cellulose synthase-like F6 ( CSLF6 ), a crucial cell wall biosynthetic gene, in Brachypodium and sorghum. Our analyses uncover regulators of CSLF6 and also shed light on the evolutionary conservation or divergence of gene regulatory elements among grass species. These results highlight NEEDLE’s capability to provide biologically relevant TF predictions and demonstrate its value for non-model plant species with dynamic transcriptome datasets.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

Challenges in predicting protein-protein interactions of understudied viruses: Arenavirus-human interactions

Understanding protein-protein interactions (PPIs) between viruses and host organisms is crucial for uncovering infection mechanisms and identifying potential therapeutic targets. The ability to generalize PPI predictive models across understudied viruses presents a significant challenge. In this work, we use arenavirus-human PPIs to illustrate the difficulties associated with model generalization, which are compounded by a lack of both positive and negative data. We employ a Transfer Learning approach to investigate arenavirus-human PPIs by utilizing models trained on better-studied virus-human and human-human PPIs. Additionally, we curate and assess four types of negative sampling datasets to evaluate their impact on model performance. Despite the overall high accuracies (93–99 %) and AUPRC scores (0.8–0.9) appearing promising, further analysis indicates that these performance metrics can be misleading due to data leakage, data bias, and overfitting, especially concerning under-represented viral proteins. We reveal these gaps and assess the impact of data imbalance using standard k-fold cross-validation and Independent Blind Testing with a Balanced Dataset, resulting in a drop in accuracy below 50 %. We propose a viral protein-specific evaluation framework that categorizes viral proteins into majority and minority classes based on their representation in the dataset, enabling comparison of model performance across these groups using balanced accuracies. This framework offers a more robust evaluation of model generalizability, addressing biases inherent in standard evaluation techniques and paving the way for more reliable PPI prediction models for understudied viruses.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Emulation and detection of physical faults and cyber-attacks on building energy systems through real-time hardware-in-the-loop experiments

The increasing use of remote or mobile access, integrated wearable technologies, data exchange, and cloud-based data analytics in modern smart buildings is steering the building industry towards open communication technologies. The increased connectivity and accessibility could lead to more cyber-attacks in smart buildings. On the other hand, physical faults (e.g., HVAC -heating, ventilation, and air-conditioning faults) may have similar adverse impacts as those from the cyber-attacks on building energy systems, such as occupant discomfort, energy wastage, and equipment downtime. However, current physical behavior-based anomaly detection methods fail to differentiate between cyber-attacks and physical faults in building energy systems. Moreover, the challenge in collecting real-world threat data with ground truth has led researchers to rely on numerical models with user-defined assumptions, which may not accurately reflect real-world conditions due to the lack of in-situ experimental datasets. To address these challenges and gaps, this paper presents a flexible hardware-in-the-loop (HIL) testbed for generating cyber-attack and physical fault datasets and demonstrating threat detection algorithms in a real building automation system (BAS) environment. This testbed combines hardware (i.e., real BAS with local HVAC controllers and a physical network) with software (i.e., high-fidelity models to represent behaviors of building envelope and HVAC energy systems), enabling emulations of realistic threats. Five HIL experiments, including one baseline without any threats, two with physical faults, and two with cyber-attacks, were conducted to generate datasets containing detailed network traffic and system states. A joint classification framework, incorporating a network analyzer and a physical HVAC fault detector, was proposed to automatically detect cyber-physical abnormalities on BAS at both the network and the physical HVAC levels. The network analyzer comprises a conditional random fields (CRF) based command validator and a statistics-based detection strategy. The fault detector employs a weather and schedule-based pattern matching and feature-based principal component analysis (WPM-FPCA) method. Evaluation of the classification using four metrics from the multi-class confusion matrix revealed an average accuracy of 90.2%, recall of 89.7%, precision of 88.5% and F1-score of 89.2%. Finally, these results demonstrate that the proposed joint classification framework can effectively differentiate between specific types of cyber-attacks (e.g., device reinitialization attack, network Denial-of-Service attack) and physical faults (e.g., air handling unit operational fault, cooling coil valve stuck) in real time for improved building energy management.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Predicting the creep-rupture lifetime of a cast austenitic stainless steel using Larson-Miller and Wilshire parametric approaches

An experimental dataset of just over 100 creep tests of a cast austenitic stainless steel, CF8C-Plus, was analyzed by two temperature-compensated parametric models (Larson-Miller, Wilshire et al.) to predict long-term lifetimes as functions of temperature and stress. The dataset and associated regression analyses showed greater scatter than typically found in recent similar studies of wrought Ni-based alloys by the same two models and was attributed to the microstructural inhomogeneity of the cast stainless steel. Qualitatively, the Larson-Miller formalism showed greater lifetime prediction accuracy than the Wilshire approach, with the latter model's predictive ability being particularly degraded by the presence of two very significant outlier results. This observation suggests that the Larson-Miller approach is more robust when treating rupture-time datasets that show particularly wide experimental scatter. Despite the differences in the overall predictive ability, both models yielded similar predictions of the applied stress at which CF8C-Plus would have a creep-limited lifetime of 100,000 h when loaded below the yield point.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Meta-analysis of biogas upgrading to renewable natural gas through biological CO 2 conversion

Biogas upgrading through CO 2 conversion by hydrogenotrophic methanogenesis is receiving an increasing attention worldwide because of the demand for renewable natural gas. Herein, a holistic and statistical study of the operation conditions, driving forces, performances, and potential implementation of biogas upgrading via biological CO 2 conversion was conducted. Based on a systematic review and meta-analysis of 46 existing publications that were selected from 1475 papers, we have compiled a global dataset of CO 2 bioconversion biogas upgrading, encompassing 308 study cases. Subsequently, we employed a rigorous analytical framework incorporating data processing and mixed effects linear regression analysis to examine the dataset. This analysis revealed a significant positive relationship between the H 2 :CO 2 ratio and the methane percentage in the upgraded biogas when using the study as a random effect. Furthermore, we performed meta-analysis on observations taken when the ratio was close to 4:1 and found that ex situ reactors (91.93% [88.11%, 95.75%]) can perform better than in situ reactors (84.74% [80.69%, 88.80%]). No evidence of differential performance was found based on the present dataset between different temperature regimes or operation modes. Furthermore, those findings establish a database that will contribute to a deeper understanding of the biogas upgrading via biological CO 2 conversion.

Biogas upgrading↗
DOE OSTI2022

The Global LAnd Surface Satellite (GLASS) evapotranspiration product Version 5.0: Algorithm development and preliminary validation

An accurate estimation of spatially and temporally continuous global terrestrial evapotranspiration (ET) is essential in the assessment of surface energy, water and carbon cycles. The Global LAnd Surface Satellite (GLASS) ET product Version 4.0 (v4.0) based on the Bayesian model averaging (BMA) method was generated to estimate global terrestrial ET. However, certain uncertainty for the GLASS ET product v4.0 limits its application. In this study, we introduced the deep neural networks (DNN) merging framework to improve terrestrial ET estimation for GLASS ET product Version 5.0 (v5.0) generation by integrating five satellite-derived ET products [Moderate Resolution Imaging Spectroradiometer (MODIS) ET product (MOD16), Shuttleworth–Wallace dual-source ET product (SW), Priestley–Taylor-based ET product (PT-JPL), modified satellite-based Priestley–Taylor ET product (MS-PT) and simple hybrid ET product (SIM)]. We compared the performance of DNN method against other merging methods, including GLASS ET algorithm v4.0 (BMA), the gradient boosting regression tree (GBRT) method and the random forest (RF) method, based on 195 global eddy covariance (EC) flux towers covering observations from 2000 through 2015. Validations indicated that the DNN had the highest accuracy among four merging methods across different land cover types, yielding the highest average determination coefficients (R 2 , 0.62), root-mean-squared-error (RMSE, 24.1 W/m 2 ) and Kling–Gupta efficiency (KGE, 0.77) with a of 99% confidence interval. Compared with GLASS ET algorithm v4.0, the DNN improved on the R 2 by approximately 7% (p < 0.01) and the KGE by 10%. Based on the DNN, we then generated 8-day GLASS ET product v5.0 globally with a 1 km spatial resolution from 2001 to 2015 driven by GLASS vegetation and surface net radiation (R n ) datasets and Modern-Era Retrospective Analysis for Research and Applications, Version 2 (MERRA2) datasets. Finally, this global terrestrial ET product provides a valuable dataset for monitoring regional and global water resources and environmental changes.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Influence of sampling frequency and estimation method on phosphorus load uncertainty in the Western Lake Erie Basin, Ohio, USA

Accurate estimates of nutrient loads are necessary to identify critical source areas and quantify the impact of management practices on pollutant export. Previous studies have investigated nutrient load estimate uncertainty, but they often focus on nutrient loads estimated using an interpolation method for large-scale watersheds with short-term datasets. The study objective was to quantify uncertainty in soluble reactive phosphorus (SRP), total phosphorus (TP), and suspended solids (SS) load estimates from two small (<10 3 km 2 ) agricultural watersheds in the western Lake Erie Basin resulting from different sampling frequencies. Each watershed had high temporal resolution datasets of discharge (15 min) and nutrient concentration (1 to 3 samples per day) collected over a 30-year period (1990–2020). Firstly, SRP, TP, and SS loads were calculated using the high temporal resolution datasets, which was assumed as “true loads”. Secondly, the high temporal concentration data were decomposed to semiweekly, weekly, biweekly, and monthly sampling and annual loads were estimated using four common load estimation methods to assess the effect of sampling frequency and load estimation method on load estimate error. Across the four different methods, the composite method had the lowest relative root mean square and absolute bias, but the rectangular interpolation method was the most precise. Furthermore, even with semiweekly sampling, the composite method resulted in an unacceptable level of precision (average imprecision = 39 %), while the interpolation method resulted in an unacceptable bias (average absolute bias = 16 %). Because neither method could provide acceptable accuracy and precision at the lowest decrease in sampling (e.t. semiweekly sampling), continued daily sampling is recommended in these watersheds.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Multi-fidelity Fourier neural operator for fast modeling of large-scale geological carbon storage

Deep learning-based surrogate models have been widely applied in geological carbon storage (GCS) problems to accelerate the prediction of reservoir pressure and CO2 plume migration. Large amounts of data from physics-based numerical simulators are required to train a model to accurately predict the complex physical behaviors associated with this process. In practice, the available training data are always limited in large-scale 3D problems due to the high computational cost. Therefore, we propose to use a multi-fidelity Fourier neural operator (FNO) to solve large-scale GCS problems with more affordable multi-fidelity training datasets. FNO has a desirable grid-invariant property, which simplifies the transfer learning procedure between datasets with different discretization. Here, we first test the model efficacy on a GCS reservoir model being discretized into 110 k grid cells. The multi-fidelity model can predict with accuracy comparable to a high-fidelity model trained with the same amount of high-fidelity data with 81% less data generation costs. We further test the generalizability of the multi-fidelity model on a same reservoir model with a finer discretization of 1 million grid cells. This case was made more challenging by employing high-fidelity and low-fidelity datasets generated by different geostatistical models and reservoir simulators. We observe that the multi-fidelity FNO model can predict pressure fields with reasonable accuracy even when the high-fidelity data are extremely limited. The findings of this study can help for better understanding of the transferability of multi-fidelity deep learning surrogate models.

58 GEOSCIENCES↗