Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Data Extrapolation”

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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Metamodeling Time-Temperature Creep Parameters

Abstract There exist many time-temperature parameter (TTP) models for creep rupture prediction of components including the Larson–Miller (LM), Manson–Haferd (MH), Manson–Brown (MB), Orr–Sherby–Dorn (OSD), Manson–Succop (MS), Graham–Walles (GW), Chitty–Duval (CD), Goldhoff–Sherby (GS) models. It remains a challenge to determine which model is “best”, capable of accurate interpolation and physically realistic extrapolation of creep rupture data for a given material. In this study, metamodeling is applied to create a unified TTP metamodel that combines and regresses into twelve TTP models (eight existing and four newly derived). An analysis of the mathematical problems that exist in TTP models is provided. A matlab code is written that can: (1) calibrate the material constants of any of the twelve TTP models (using the metamodel); (2) determine the most suitable stress-parameter function; (3) and report the normalized mean square error (NMSE) of rupture predictions for a given material database. Using the metamodel, and code, a design engineer can make an intelligent selection of the “best” TTP model for creep resistant design. This process is demonstrated using four isotherms of alloy P91 creep rupture data. To assess the influence of material, further validation is performed on alloys Hastelloy X, 304SS, and 316SS. It is determined that the “best” model is dependent on material type and the quality and quantity of available data.

Engineering↗
DOE OSTI2020

Implications of Non-Bingham Rheology

One of the key challenges now facing the US Department of Energy (DOE) is the fate of radioactive waste remaining from World War II and the Cold War which is stored underground in tanks some 75 feet in diameter and over 30 feet tall. Over time, the waste has segmented into multiple layers with sludges and slurries at the bottom with salt crust layers often at the top and liquid in between. DOE’s current official baseline plans call for remaining sludges and slurries to be removed from the tanks and converted into a more stable glass waste form. Minimizing worker exposure to radiation drives DOE to use slurry processing techniques to suspend, mobilize, transport, mix, and process the waste. Therefore, a clear and quantitative understanding of Hanford waste rheology is essential for the success of the DOE mission. Historically much of the waste has been characterized using Eugene Bingham’s model more than a century old that provides a straight line fit to higher shear rate data with the intercept suggesting a yield stress and the slope providing the consistency. Yet, Bingham fits overestimate the shear stress at a given shear rate for low to intermediate shear rates, exactly the range of shear rates typically encountered in pipe flow, where shear rates peak along the pipe wall and vanish in the center. This model produces a fictitious yield stress for some of the wastes that do not exhibit yield phenomena. While over estimating the yield stress may be prudent, safe, and conservative for some applications (e.g., pump sizing to ensure that pumps can handle yield stresses), over estimating the rheology may be inaccurate and non-conservative for other applications (e.g., eroding settled particle beds). Therefore, this paper evaluates the slurry rheology of Hanford and Savannah River wastes using a more modern rheological model that fits the full range of experimental data. Although a bias has been recognized and alternative models proposed, the magnitude of this bias and the implications for tank waste have only been qualitatively suggested. The purpose of this paper is to evaluate quantitatively implications of the poor quality of fit between a Bingham model for rheology and experimental data at modest shear rates. We first demonstrate the magnitude of the bias between the data and the Bingham extrapolation. We then evaluate quantitatively the velocity profile under laminar conditions. This analysis shows that the bias may be large (hundreds of percent or more) at modest shear rates and that modest shear rates dominate pipe velocity profiles.

Rheology, Radioactive waste↗
DOE OSTI2021

Paying it Forward: How an Emerging Unconventional Play can Hit the Ground Running

While every tight oil play is unique, there are lessons that can be transferred from one play to another to improve the efficiency and pace of production operations and development. These improvements may not fit precisely in every basin or play but generally hold to themes that can be tested against and built upon. Themes such as the quantity of proppant, longer lateral length, or the number of stages can be directly tied to increased productivity. However, there are diminishing returns on these investment activities for which each operator, within a given play, will be required to identify and mitigate against. This is especially true as the industry steps in and begins developing new tight oil plays. In their nascent stages, these plays may have limited well penetrations and, as a result, limited geological and performance data from which to extrapolate. Pulling together an understanding of where the industry currently resides in terms of how to exploit these resources can provide a boost in terms of working towards greatly improved well completions. In 2019, the US EIA estimated that nearly 8 million barrels of oil per day were produced from tight oil reservoirs in the United States (US EIA, 2020). This represents over 60% of the domestic crude production, originating from multiple reservoirs in the Permian Basin (TX) as well as the Bakken (MT, ND), Eagle Ford (TX), Niobrara (CO, WY), and Anadarko Basin (OK) formations, among others. As such, there are 1,000s of wells across these numerous tight oil plays that can relate an informative story. To build this story, the interplay of geology, well spacing, lateral length, and stimulation, all critical to economic success, will be explored. This paper proposes to look back at these mature tight oil (and gas) basins and bring forth an understanding of what lessons can be applied to the emerging Powder River Basin tight oil reservoirs (Mowry and the Turner/Frontier). The authors will delve into the four broad topics of geology, well spacing, lateral length, and stimulation, highlighting case studies to demonstrate those lessons from established tight oil plays that will underpin planned activities at a Field Laboratory Test Site in the southern Powder River Basin.

02 PETROLEUM↗
DOE OSTI2023

Biochemical Conversion of Lignocellulosic Biomass to Hydrocarbon Fuels and Products: 2022 State of Technology and Future Research

The annual State of Technology (SOT) assessment is a key activity for biochemical platform research. It allows the impact of research progress (both directly achieved in-house at the National Renewable Energy Laboratory [NREL] and indirectly extrapolated from available public data) to be quantified in terms of economic improvements in the overall cellulosic biofuel production process for a particular conversion pathway. As such, initial benchmarks can be established for currently demonstrated performance and progress can be tracked toward out-year goals to ultimately demonstrate cost-competitive biofuel technology. Building upon efforts to progress NREL's biochemical platform R&D work toward ultimate 2030 goals to demonstrate less than $2.50/gallon gasoline equivalent (GGE) fuel selling prices, experimental and techno-economic analysis (TEA) activities have primarily focused on "advanced" biochemical processing strategies to fuels and coproducts, guided by TEA modeling to highlight key barriers and priorities toward achieving this goal across a number of potential bioconversion pathways. The purpose of the present effort is to benchmark the latest experimental developments for these pathways as quantified by modeled minimum fuel selling prices (MFSPs), as a measure of current status relative to those final targets. For this SOT, TEA models were run for two separate biological conversion pathways to fuels, based on available data for integrated biomass deconstruction and hydrolysate processing; namely carboxylic acids (primarily butyric acid) and diols (2,3-butanediol [BDO]), reflecting NREL's recently published 2018 biochemical design report focused on those two pathways. The models were run across two scenarios for lignin utilization, namely combustion and conversion to value-added coproducts.

09 BIOMASS FUELS↗
DOE OSTI2023

Carbon Fiber Reinforced Plastic with Substantially Improved Through-plane Thermal Conductivity

Thermal conductivity of carbon fiber support structures that also serve as cooling substrates is of increasing importance for high energy physics detectors. One of the shortcomings of current carbon fiber laminates is that they exhibit excellent thermal conductivity of order several hundred W/(m*K) along the fiber direction, whereas the through plane thermal conductivity perpendicular to the fibers is orders of magnitude lower, and at about 1 W/(m*K) more in the range of thermal insulators. This proposal aims to significantly improve the through plane thermal conductivity, by at least a factor five, changing CFRP from being a thermal insulator to a thermal conductor in the through plane direction. Between June 2021 and December 2022, a total of 36 carbon fiber laminate test samples and two carbon loaded epoxy test samples were manufactured and analyzed for thermal performance. Unfortunately, the promising results obtained with a similar sample manufactured in 2017 were not reproduced. We believe this to be mainly due the excessive age of the carbon fiber prepreg material used for this SBIR - the same material batch, produced in 2015, was used in 2017 and in 2021/2022, i.e. the material was seven years old when the final samples were produced - , and to a lesser degree to the difference in pressure attainable with the 2017 and 2021/2022 laminate curing setups. While disappointing, the results are not completely discouraging, since for the samples with the highest compression the through-plane thermal conductivities plotted as a function of per-ply thickness extrapolate well to the data point from the 2017 sample.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Ion Exchange Processing of AN-107 Hanford Tank Waste through Crystalline Silicotitanate in a Staged 2- then 3-Column System

The Hanford Site stores an estimated 56 million gallons of mixed radioactive and chemically hazardous waste in large underground tanks. In support of the Direct Feed Low-Activity Waste (DFLAW) Program for expediting Hanford tank waste supernate treatment, laboratory-scale ion exchange processing using prototypic unit operations was conducted on AN-107 tank waste at the Pacific Northwest National Laboratory Radiochemical Processing Laboratory. This report describes the small-scale ion exchange testing with 13.7 L of diluted and filtered supernate from Tank 241-AN-107 (hereafter referred to as AN-107) at 16 °C (62 °F). One of the waste acceptance criteria (WAC) for the Waste Treatment Plant (WTP) Low-Activity Waste Facility is that the waste must contain less than 3.18×10 -5 Ci 137 Cs per mole of Na. For the AN-107 tank waste to meet this criterion, only 0.147% of the influent 137 Cs concentration may be delivered to the WTP; this requires a Cs decontamination factor of 678. Testing with AN-107 matched current Tank Side Cesium Removal (TSCR) facility prototypic operations where a lead-lag configuration was used until the lag column reached the WAC limit, then a polish column was brought online for continued processing in a lead-lag-polish column configuration. Feed was processed at 1.9 bed volumes (BVs) per hour; the flowrate, in terms of contact time with the crystalline silicotitanate (CST) bed, matched the expected flowrate at TSCR. The Cs-decontaminated product was retained for vitrification testing (to be reported separately). The lead column reached 40% Cs breakthrough after processing ~1700 BVs of feed; the 50% Cs breakthrough was extrapolated from the breakthrough data to occur at 1873 BVs. Testing compared to previous AP-101 and AP-107 testing at 16 °C showed ~300 BV increases in volume processed to reach the WAC limit for both lead and lag columns. The increase in capacity was determined to be due to the significantly lower K concentration in the AN-107 compared to the other tank waste matrices. A comparison in breakthrough curves for the three tests indicated slightly slower kinetic behavior in the AN-107, with variations in feed matrices (high organic complexants) likely responsible for the deviation. The Cs effluent from the lag column reached the WAC limit after processing 1097 BVs. Anticipating this breakthrough point, the polish column was preemptively installed around 900 BVs. Cs breakthrough from the lag column began at 500 BVs, reaching 3.06×10 0 µCi/mL, or 2.6 % Cs breakthrough, after processing all 1700 BVs of feed. Table S.1 and Figure S.1 summarize the observed column performance and relevant Cs loading characteristics.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2024

Carbon Storage Core Characterization Efforts at NETL

The multi-scale Computed Tomography (CT) and core flow facility in the Geocharacterization Laboratory at NETL, Morgantown yields porosity, permeability, and fracture properties of rock core samples obtained from the subsurface while maintaining the integrity of the sample. Additionally, geophysical bulk rock properties are analyzed with the laboratory’s GeoTEK multi-sensor core logger in a comparable fashion to downhole methods. NETL researchers collaborate with stakeholders within the carbon storage, oil and gas, and critical minerals sectors. Since 2017, over 1.88 miles of core have been analyzed within the laboratory and all data is publicly available through the Technical Report Series (TRS) on the Energy Data eXchange (EDX). Additionally, the website, RokBase, was curated to extrapolate and visualize the high-resolution data from field operations. The characterization of the Lively Grove #1 (LG#1) well provides a case study into the full capabilities of the Geocharacterization Laboratory. During the comprehensive study of LG#1 ~1-2 mm in diameter, vertical to bedding, cylindrical structures were identified throughout the St. Peter Formation. These structures are pervasive throughout the St. Peter Formation at depth and are characterized as the trace fossil, Skolithos.

Isom, Shelby L↗
DOE OSTI2023

Uncertainty quantification of machine learning models to improve streamflow prediction under changing climate and environmental conditions

Machine learning (ML) models, and Long Short-Term Memory (LSTM) networks in particular, have demonstrated remarkable performance in streamflow prediction and are increasingly being used by the hydrological research community. However, most of these applications do not include uncertainty quantification (UQ). ML models are data driven and can suffer from large extrapolation errors when applied to changing climate/environmental conditions. UQ is required to quantify the influence of data noises on model predictions and avoid overconfident projections in extrapolation. In this work, we integrate a novel UQ method, called PI3NN, with LSTM networks for streamflow prediction. PI3NN calculates Prediction Intervals by training 3 Neural Networks. It can precisely quantify the predictive uncertainty caused by the data noise and identify out-of-distribution (OOD) data in a non-stationary condition to avoid overconfident predictions. We apply the PI3NN-LSTM method in the snow-dominant East River Watershed in the western US and in the rain-driven Walker Branch Watershed in the southeastern US. Results indicate that for the prediction data which have similar features as the training data, PI3NN precisely quantifies the predictive uncertainty with the desired confidence level; and for the OOD data where the LSTM network fails to make accurate predictions, PI3NN produces a reasonably large uncertainty indicating that the results are not trustworthy and should avoid overconfidence. PI3NN is computationally efficient, robust in performance, and generalizable to various network structures and data with no distributional assumptions. It can be broadly applied in ML-based hydrological simulations for credible prediction.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2020

Measurement of the scintillation and ionization response of liquid xenon at MeV energies in the EXO-200 experiment

In this work, liquid xenon (LXe) is employed in a number of current and future detectors for rare event searches. We use the EXO-200 experimental data to measure the absolute scintillation and ionization yields generated by γ interactions from 228 Th (2615 keV), 226 Ra (1764 keV), and 60 Co (1332 keV and 1173 keV) calibration sources, over a range of electric fields. The W value that defines the recombination-independent energy scale is measured to be 11.5±0.5 (syst.) ±0.1 (stat.) eV. These data are also used to measure the recombination fluctuations in the number of electrons and photons produced by the calibration sources at the MeV scale, which deviate from extrapolations of lower-energy data. Additionally, a semiempirical model for the energy resolution of the detector is developed, which is used to constrain the recombination efficiency, i.e., the fraction of recombined electrons that result in the emission of a detectable photon. Detailed measurements of the absolute charge and light yields for MeV-scale electron recoils are important for predicting the performance of future neutrinoless double β -decay detectors.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

Deep-learning-based canopy height model generation from sub-meter resolution panchromatic satellite imagery

Canopy height models (CHMs) with sufficient resolution to distinguish individual trees are useful for a variety of applications. However, standard techniques to acquire such data, such as airborne lidar surveying, are often prohibitively expensive. Deep learning techniques for generating CHMs from high-resolution imagery are an attractive option to reduce costs. To date, success with these methods has been demonstrated using multichannel aerial photography and specialized satellite data products derived from multiple sensors, neither of which is commonly available at temporal resolutions finer than one year. Here we demonstrate a method to generate sub-meter resolution CHMs in three forests in California using a more abundant data source: sub-meter resolution, panchromatic satellite imagery from a single sensor. We show that phenology and species composition play important roles in model transferability; when trained using imagery from a single conifer forest in autumn, the model performs well on autumn imagery from a second conifer forest several hundred kilometers distant with no re-training. With modest additions to the training dataset, the same model generates minimally biased estimates of canopy height in both conifer and deciduous forests during multiple seasons. Because the model operates on satellite data with global coverage and a relatively short return interval, we propose its suitability to extrapolate tree-level canopy height data to remote regions and conduct high-temporal resolution monitoring of forest structure. We furthermore demonstrate the workflow’s applicability to fire modeling by conducting simulations in forests populated by trees measured using both this approach and airborne lidar surveying. We find minimal differences in fire behavior relative to a baseline case in which only statistical distributions of tree height and crown area are known. This result underscores the value of forest structural information derived from our workflow for improving the fidelity of wildland fire simulations, among other ecological applications.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

An initial framework for the rapid qualification of long-term creep rupture strength via microstructural modeling

This report describes the development and testing of a new method for extrapolating short-term creep rupture test data to predict long-term rupture strength. The goal of this work is to reduce the time required to qualify new materials for nuclear service by reducing the lead time required for dedicated, long-term material testing to establish key long-term material properties. The new approach described here uses a physics-based model to predict the long-term creep rupture strength of 316H stainless steel using only short-term test data. The key idea is to use Bayesian inference to find the statistical distribution of the model parameters that best explain the short-term rupture data. Because the model is physics-based these parameters are all microstructural quantities that can be measured through detailed material characterization experiments. The Bayesian prior distributions provide a means for incorporating this characterization data into the final model to improve the accuracy of the long-term model predictions. However, where such data is not available the process still produces an accurate model based on an uniformed prior. Our hypothesis is that this approach more accurately extrapolates the short-term test data when compared to current, empirical methods. The report proves this hypothesis using actual long-term rupture data available for 316H, including tests with rupture times greater than 200,000 hours. The general approach developed here could be applied to other materials and other time-dependent material properties. Applying this new technique to develop long-term qualified material properties, potentially in conjunction with other accelerated qualification approaches like staggered qualification test programs, could greatly reduce the time required to qualify new materials for nuclear service.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2026

Added value of site load measurements in probabilistic lifetime extension: a Lillgrund case study

Site-specific fatigue estimation is an essential part of wind turbine lifetime extension, with various methods depending on data availability. The present study compares probabilistic lifetime extension assessment results for rotor blades with and without load measurements. It also addresses two key questions in such assessments: the applicability of the Frandsen model for estimating waked turbulence under complex and mixed wake conditions and the extrapolation of mid-term data over longer time periods. The case study wind turbine is SWT-2.3-93, located at the edge of the Lillgrund wind farm, situated in the Øresund Strait between Denmark and Sweden. The turbine is extensively instrumented, with 5 years of data available from its supervisory control and data acquisition (SCADA) system. Although the Frandsen turbulence estimates deviate in a different manner from measurements at below- and above-rated mean wind speeds, the model remains a conservative approach for fatigue load prediction and reliability. In the current case study, the site-specific assessment using strain gauge measurements yields a 33 % higher annual fatigue reliability index after 35 years compared to a scenario based on the Frandsen estimation combined with ambient environmental data and a generic aeroelastic model. The results also demonstrate that the sensitivity of fatigue reliability to load uncertainty is negligible when load measurements are used directly but relatively high when relying on the Frandsen model in combination with a generic aeroelastic model. Overall, the high variability of the lifetime extension in different scenarios of data availability and accuracy shows the importance and added value of high-quality measurements combined with wind-farm-level SCADA and a model updated in real time (digital twins).

17 WIND ENERGY↗
DOE OSTI2024

$\overline{TKE}$ Parameterization and $\bar{v}$ Uncertainty Analysis for CGMF

Previous work was performed on tuning CGMF parameters for 235 U, 238 U, and Plutonium isotopes. Now work is being done to tune minor uranium isotopes. However, uranium isotopes like 232 U and 236 U have almost no experimental data. We are applying cross-isotope models to extrapolate and tune CGMF on isotopes that lack experimental data. There exist several internal CGMF physics quantities that affect the output of CGMF—multi-chance fission probability, excitation energy sharing, spin-cutoff factor, spin scaling, and fragment total kinetic energy to name a few. The mean fragment total kinetic energy, $\overline{TKE}$, is particularly interesting because of its strong anti-correlation with $\bar{v}$. We are most interested in the mean fragment total kinetic energy before neutron emissions. $\overline{TKE}$ is assumed to be pre-neutron emission unless otherwise stated. Currently in CGMF, the $\overline{TKE}$ model for 233,234,235,238 U are tuned independently to reproduce ν for the associated isotopes. In this report, we will tune a cross-isotope $\overline{TKE}$ model to experimental $\overline{TKE}$ data for 232,233,234,235,236,238 U. Because of the unreliable and sparse nature of $\overline{TKE}$ experimental data, future work will use more reliable experimental $\bar{v}$ data to infer the $\overline{TKE}$ model (and likely other internal CGMF parameters) for uranium isotopes. Such work has been performed previously using a sensitivity analysis and Kalman filter methods.

07 ISOTOPE AND RADIATION SOURCES↗
DOE OSTI2026

PISCES two-detector covariance matrix fit for the NOvA Experiment

NOvA is a long-baseline neutrino oscillation experiment with two functionally identical detectors: a Near Detector (ND) at Fermilab, placed 1 km from the neutrino source, and a Far Detector (FD) located 810 km away from the ND in Minnesota. NOvA's primary physics goals are the precise measurements of neutrino oscillation parameters $\theta_{23}$ and $\Delta m^2_{32}$ , determine the neutrino mass ordering, and constrain the value of $\delta_{CP}$, via the study of muon neutrino to electron neutrino oscillation. In the standard NOvA three-flavor analysis, oscillation parameters are extracted using an extrapolation technique in which the ND data constrain the FD prediction through a ratio method. While this allows for systematic uncertainties sharing the same effects in both detectors to cancel, it remains an FD-only fit and does not fully leverage the constraining power of the high-statistics ND. This analysis proposes a simultaneous ND+FD fit using the PISCES method. PISCES (Parameter Inference with Systematic Covariance and Exact Statistics) is a framework designed to support complex configurations such as a joint ND+FD fit. This allows PISCES to take full advantage of the ND data to directly constrain systematic uncertainties across all samples. In PISCES, systematic uncertainties are encoded in a fractional covariance matrix, and statistical uncertainties are handled with a Poisson likelihood, making the approach well suited for low-statistics samples. For interpretability, we further use a Newton–Raphson + PCA method to recover per-systematic pulls from the covariance formulation. This poster presents the full PISCES joint ND+FD fit for the NOvA three-flavor analysis, describes its implementation and evaluates its performance through extensive robustness tests and fake data studies. It also provides a comparison between the PISCES joint ND+FD results and the standard NOvA extrapolation method.

Rajaoalisoa, Miriama [Cincinnati U.] (ORCID:000000↗
DOE OSTI2026

Measurements of three-flavor neutrino oscillations from a PISCES two-detector fit to the NOvA Experiment data

NOvA is a long-baseline neutrino oscillation experiment with two functionally identical detectors: a Near Detector (ND) at Fermilab, placed 1 km from the neutrino source, and a Far Detector (FD) located 810 km away from the ND in Minnesota. NOvA s primary physics goals are to measure the neutrino oscillation parameters $\theta_{23}$ and $\Delta m^2_{32}$ with high precision, determine the neutrino mass hierarchy, and constrain the value of $\delta_{CP}$, primarily via the study of muon neutrino to electron neutrino oscillation. Extracting values for oscillation parameters from fits to data usually relies on treating systematic uncertainties as nuisance parameters, a strategy that suffers from poor scalability as the number of uncertainties becomes larger. This work introduces PISCES (Parameter Inference with Systematic Covariance and Exact Statistics), a novel method that circumvents this scalability problem by encoding systematic uncertainties into a covariance matrix. PISCES utilizes a nested minimization in which optimal systematic pulls are first computed using the covariance matrix in an inner minimization step, then the oscillation parameters are profiled over in the outer minimization. PISCES also uses a Poisson Likelihood term, making it ideal for the inclusion of low-statistic samples in the fits. PISCES is a flexible framework that also supports complex fits, such as a joint Near and Far detector fit. In the standard NOvA analysis, oscillation parameters are extracted using an extrapolation technique in which the ND data indirectly constrain the FD prediction via a ratio method. PISCES, on the other hand, enables a simultaneous ND+FD fit, allowing the high-statistics ND data to directly constrain systematic uncertainties across all samples. This thesis presents the full PISCES joint ND+FD fit for the NOvA three-flavor analysis, details its implementation, and evaluates its performance through extensive robustness tests and fake data studies. It also provides a comparison between the PISCES joint ND+FD results and the standard NOvA extrapolation method using the full NOvA 10-year data set. The results demonstrate that PISCES can successfully fit NOvA data while incorporating the constraints from the ND detectors consistently, using physically motivated systematic uncertainties to account for data/MC discrepancies.

Rajaoalisoa, Miriama [Cincinnati U.]↗
DOE OSTI2023

Extrapolation of the Rainflow-Counted Load Ranges for Fatigue Assessment of the Wind Turbine's Blades

Wind turbine design standards recommend the use of statistical modeling coupled with extrapolation of the short-term load data to long-term periods for fatigue reliability assessment. However, statistical error and computational expense can limit the accuracy of such approaches. In the case of wind turbine blades, the errors are more significant because of the high material fatigue exponent that makes the damage estimations more sensitive to variations. In addition, due to different excitation sources, the flapwise load range histogram is not unimodal, and thus its statistical modeling is complex. In the present work, we provide three methods for statistical modeling of the flapwise bending moment ranges including a novel approach based on frequency-based separation of the modes. The first two methods are simplified approaches for modeling the most crucial load ranges using unimodal distributions and the third method involves multimodal distribution fitting. The research is based on 3600 10-minute aeroelastic simulations of DTU 10MW case study wind turbine from which a benchmark damage equivalent load (DEL) is calculated. The DEL calculated by each of the three proposed methods is compared to this reference. The results show that the conventional approach based on using 6 seeds as well as using mixture models fitted on the limited data lead to under-conservative results with errors up to 23%. On the other hand, the simplified unimodal approaches provided in this work can provide conservative estimations of the fatigue damage with mean values 5% and 12% higher than the benchmark. However, the variability of the DEL estimates is higher when using unimodal extrapolation of the load ranges, and the data can be conservative by 17.5%. The proposed unimodal fits suggested for modeling and extrapolation of the blade's load ranges provide less errors relatively and most importantly conservative DEL estimations while maintaining computational efficiency.

blade fatigue↗
DOE OSTI2021

Connect the Dots: In Situ 4-D Seismic Monitoring of CO 2 Storage With Spatio-Temporal CNNs

4-D seismic imaging has been widely used in CO 2 sequestration projects to monitor the fluid flow in the volumetric subsurface region that is not sampled by wells. Ideally, real-time monitoring and near-future forecasting would provide site operators with great insights to understand the dynamics of the subsurface reservoir and assess any potential risks. However, due to obstacles such as high deployment cost, availability of acquisition equipment, exclusion zones around surface structures, only very sparse seismic imaging data can be obtained during monitoring. That leads to an unavoidable and growing knowledge gap over time. The operator needs to understand the fluid flow throughout the project lifetime and the seismic data are only available at a limited number of times. This is insufficient for understanding reservoir behavior. To overcome those challenges, we have developed spatio-temporal neural-network-based models that can produce high-fidelity interpolated or extrapolated images effectively and efficiently. Specifically, our models are built on an autoencoder, and incorporate the long short-term memory (LSTM) structure with a new loss function regularized by optical flow. We validate the performance of our models using real 4-D post-stack seismic imaging data acquired at the Sleipner CO 2 sequestration field. We employ two different strategies in evaluating our models. Numerically, we compare our models with different baseline approaches using classic pixel-based metrics. We also conduct a blind survey and collect a total of 20 responses from domain experts to evaluate the quality of data generated by our models. Finally, via both numerical and expert evaluation, we conclude that our models can produce high-quality 2-D/3-D seismic imaging data at a reasonable cost, offering the possibility of real-time monitoring or even near-future forecasting of the CO 2 storage reservoir.

4-D seismic imaging↗
DOE OSTI2022

Learning functional priors and posteriors from data and physics

In this work, we develop a new Bayesian framework based on deep neural networks to be able to extrapolate in space-time using historical data and to quantify uncertainties arising from both noisy and gappy data in physical problems. Specifically, the proposed approach has two stages: (1) prior learning and (2) posterior estimation. At the first stage, we employ the physics-informed Generative Adversarial Networks (PI-GAN) to learn a functional prior either from a prescribed function distribution, e.g., Gaussian process, or from historical data and physics. At the second stage, we employ the Hamiltonian Monte Carlo (HMC) method to estimate the posterior in the latent space of PI-GANs. In addition, we use two different approaches to encode the physics: (1) automatic differentiation, used in the physicsinformed neural networks (PINNs) for scenarios with explicitly known partial differential equations (PDEs), and (2) operator regression using the deep operator network (DeepONet) for PDE-agnostic scenarios. We then test the proposed method for (1) meta-learning for one-dimensional regression, and forward/inverse PDE problems (combined with PINNs); (2) PDE-agnostic physical problems (combined with DeepONet), e.g., fractional diffusion as well as saturated stochastic (100-dimensional) flows in heterogeneous porous media; and (3) spatial-temporal regression problems, i.e., inference of a marine riser displacement field using experimental data from the Norwegian Deepwater Programme (NDP). The results demonstrate that the proposed approach can provide accurate predictions as well as uncertainty quantification given very limited scattered and noisy data, since historical data could be available to provide informative priors. In summary, the proposed method is capable of learning flexible functional priors, e.g., both Gaussian and non-Gaussian process, and can be readily extended to big data problems by enabling mini-batch training using stochastic HMC or normalizing flows since the latent space is generally characterized as low dimensional.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗