Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “regression”

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 1,243 records · Page 69
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Shock Hugoniot calculations using on-the-fly machine learned force fields with ab initio accuracy

We present a framework for computing the shock Hugoniot using on-the-fly machine learned force field (MLFF) molecular dynamics simulations. In particular, we employ an MLFF model based on the kernel method and Bayesian linear regression to compute the free energy, atomic forces, and pressure, in conjunction with a linear regression model between the internal and free energies to compute the internal energy, with all training data generated from Kohn–Sham density functional theory (DFT). We verify the accuracy of the formalism by comparing the Hugoniot for carbon with recent Kohn–Sham DFT results in the literature. In so doing, we demonstrate that Kohn–Sham calculations for the Hugoniot can be accelerated by up to two orders of magnitude, while retaining ab initio accuracy. We apply this framework to calculate the Hugoniots of 14 materials in the FPEOS database, comprising 9 single elements and 5 compounds, between temperatures of 10 kK and 2 MK. We find good agreement with first principles results in the literature while providing tighter error bars. In addition, we confirm that the inter-element interaction in compounds decreases with temperature.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Decoding diffraction and spectroscopy data with machine learning: A tutorial

This Tutorial provides a step-by-step guide on how to apply supervised machine-learning techniques to analyze diffraction and spectroscopy data. This Tutorial details four models—a reconstruction-focused model, a regression-focused model, a hybrid reconstruction/regression model, and a multimodal model—that use x-ray diffraction profiles and vibrational density of states spectra to predict various microstructural descriptors. In this Tutorial, we cover data pre-processing steps, constructions of the models via dimensionality reduction and regression, training, and analysis of these models. Comparisons of the model’s performance are provided, highlighting the strength and weakness of the various approaches utilized.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Prediction of hydrogen adsorption in nanoporous materials from the energy distribution of adsorption sites

We present a fast and accurate, semi-analytical method for predicting hydrogen adsorption in nanoporous materials. For any temperature and pressure, the adsorbed amount is calculated as an integral over the energy density of adsorption sites (guest-host interactions) plus an average guest-guest term. The guest-host interaction energy is calculated using a classical force field with hydrogen modelled as a single-site probe. The guest-guest interaction energy is approximated using an average coordination number, which is regressed using Gaussian Process Regression (GPR). Local adsorption at each site is then modelled using a Langmuir isotherm, which when weighted with its probability density gives an accurate description of hydrogen adsorption. The method is tested on 933 metal-organic frameworks (MOFs) from the Computation-Ready Experimental (CoRE) MOF database at 77 K from 10 –5 to 100 bar, and the results are compared against GCMC predictions. To demonstrate the utility of the method, we calculated hydrogen adsorption isotherms for 12,914 existing MOF structures, at two different temperatures at a speed about 100 times that of GCMC simulations and analyzed the results. Here, we found 13 MOFs with predicted deliverable capacities exceeding the DOE target of 50 g/L for adsorption at 100 bar, 77 K and desorption at 5 bar, 160 K.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Deep Learning for Multigroup Cross-Section Representation in Two-Step Core Calculations

Here we investigate using deep learning, a type of machine-learning algorithm employing multiple layers of artificial neurons, for the mathematical representation of multigroup cross sections for use in the Griffin reactor multiphysics code for two-step deterministic neutronics calculations. A three-dimensional fuel element typical of a high-temperature gas reactor as well as a two-dimensional sodium-cooled fast reactor lattice are modeled using the Serpent Monte Carlo code, and multigroup macroscopic cross sections are generated for various state parameters to produce a training data set and a separate validation data set. A fully connected, feedforward neural network is trained using the open-source PyTorch machine-learning framework, and its accuracy is compared against the standard piecewise linear interpolation model. Additionally, we provide in this work a generic technique for propagating the cross-section model errors up to the k eff using sensitivity coefficients with the first-order uncertainty propagation rule. Quantifying the eigenvalue error due to the cross-section regression errors is especially practical for appropriately selecting the mathematical representation of the cross sections. We demonstrate that the artificial neural network model produces lower errors and therefore enables better accuracy relative to the piecewise linear model when the cross sections exhibit nonlinear dependencies; especially when a coarse grid is employed, where the errors can be halved by the artificial neural network. However, for linearly dependent multigroup cross sections as found for the sodium-cooled fast reactor case, a simpler linear regression outperforms deeper networks.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2013

WMAP OBSERVATIONS OF PLANCK ESZ CLUSTERS

We examine the Sunyaev-Zeldovich (SZ) effect in the seven year Wilkinson Microwave Anisotropy Probe (WMAP) data by cross-correlating it with the Planck Early-release Sunyaev-Zeldovich catalog. Our analysis proceeds in two parts. We first perform a stacking analysis in which the filtered WMAP data are averaged at the locations of the 175 Planck clusters. We then perform a regression analysis to compare the mean amplitude of the SZ signal, Y{sub 500}, in the WMAP data to the corresponding amplitude in the Planck data. The aggregate Planck clusters are detected in the seven year WMAP data with a signal-to-noise ratio of 16.3. In the regression analysis, we find that the SZ amplitude measurements agree to better than 25%: a = 1.23 {+-} 0.18 for the fit Y{sup wmap}{sub 500}= aY{sup planck}{sub 500}.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

A robust statistical analysis of the role of hydropower on the system electricity price and price volatility

Abstract Hydroelectric power (hydropower) is unique in that it can function as both a conventional source of electricity and as backup storage (pumped hydroelectric storage and large reservoir storage) for providing energy in times of high demand on the grid (S. Rehman, L M Al-Hadhrami, and M M Alam), (2015 Renewable and Sustainable Energy Reviews , 44 , 586–98). This study examines the impact of hydropower on system electricity price and price volatility in the region served by the New England Independent System Operator (ISONE) from 2014-2020 (ISONE, ISO New England Web Services API v1.1 .” https://webservices.iso-ne.com/docs/v1.1/ , 2021. Accessed: 2021-01-10). We perform a robust holistic analysis of the mean and quantile effects, as well as the marginal contributing effects of hydropower in the presence of solar and wind resources. First, the price data is adjusted for deterministic temporal trends, correcting for seasonal, weekend, and diurnal effects that may obscure actual representative trends in the data. Using multiple linear regression and quantile regression, we observe that hydropower contributes to a reduction in the system electricity price and price volatility. While hydropower has a weak impact on decreasing price and volatility at the mean, it has greater impact at extreme quantiles (>70th percentile). At these higher percentiles, we find that hydropower provides a stabilizing effect on price volatility in the presence of volatile resources such as wind. We conclude with a discussion of the observed relationship between hydropower and system electricity price and volatility.

13 HYDRO ENERGY↗
DOE OSTI2023

Diminishing marginal effect in estimating the dissolved organic carbon export from a watershed

Dissolved organic carbon (DOC) can be initially moved from soils to inland waters with surface runoff, and then mineralized, buried, or eventually delivered to the coastal ocean. This land-to-ocean phase of the DOC flux must be accounted for to comprehensively understand the global carbon cycle. To estimate the terrestrial-aquatic DOC leaching, calculating the product of the riverine DOC concentration and the corresponding river discharge measured at the watershed outlet is a common method. However, it is challenging to frequently and exactly record riverine DOC concentrations, thus the relationship between DOC concentrations and discharges (C-Q relationship) are established and used to interpolate the time-series of DOC concentrations. We found that the widely used time-dependent and time-independent C-Q regression models are weak in representing their altered relationship when the discharge is extremely high, which was named as diminishing marginal effect. In this study, we evaluated the performance of two C-Q regression models and discussed possible reasons for the diminishing marginal effect. We suggest that repeated and long-term measurements of the DOC concentration are required to adequately analyze their relationships, especially during the early spring and seasons with heavy precipitations.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Clarifying trust of materials property predictions using neural networks with distribution-specific uncertainty quantification

It is critical that machine learning (ML) model predictions be trustworthy for high-throughput catalyst discovery approaches. Uncertainty quantification (UQ) methods allow estimation of the trustworthiness of an ML model, but these methods have not been well explored in the field of heterogeneous catalysis. Herein, we investigate different UQ methods applied to a crystal graph convolutional neural network to predict adsorption energies of molecules on alloys from the Open Catalyst 2020 dataset, the largest existing heterogeneous catalyst dataset. We apply three UQ methods to the adsorption energy predictions, namely k-fold ensembling, Monte Carlo dropout, and evidential regression. The effectiveness of each UQ method is assessed based on accuracy, sharpness, dispersion, calibration, and tightness. Evidential regression is demonstrated to be a powerful approach for rapidly obtaining tunable, competitively trustworthy UQ estimates for heterogeneous catalysis applications when using neural networks. Recalibration of model uncertainties is shown to be essential in practical screening applications of catalysts using uncertainties.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Unifying simulation and inference with normalizing flows

There have been many applications of deep neural networks to detector calibrations and a growing number of studies that propose deep generative models as automated fast detector simulators. We show that these two tasks can be unified by using maximum likelihood estimation (MLE) from conditional generative models for energy regression. Unlike direct regression techniques, the MLE approach is prior independent and non-Gaussian resolutions can be determined from the shape of the likelihood near the maximum. Using an ATLAS-like calorimeter simulation, we demonstrate this concept in the context of calorimeter energy calibration. Published by the American Physical Society 2025

Hadronic calorimiters↗
DOE OSTI2021

Comparison of Probabilistic Forecasts for Predictive Voltage Control

This paper explores predictive cooperative voltage control in distribution systems with highly variable sources such as photovoltaics (PV). The goal is to maintain the voltage profile within the limits despite the fluctuations due to sudden changes in solar power generation. The predictive voltage control method relies on probabilistic solar power and load forecasts to select the optimum Voltage Regulator (VR) taps appropriately. VR taps are selected to minimize the risk of voltage violation. A modified version of the IEEE 123 system is used as the case study. A 100% penetration of solar power is assumed for the distribution system with profiles for solar generation and loads added to the system. Three different probabilistic forecast models (Quantile Regression (QR), Gaussian distribution and volatility forecasting using Generalized Autoregressive Conditional Heteroskedasticity (GARCH)) are explored in this study. The results for the VR taps and Voltage Deviation Index (VDI) are compared to find the most effective forecast model.

Panamtash, Hossein↗
DOE OSTI2024

Taylor-Expansion-Based Robust Power Flow in Unbalanced Distribution Systems: A Hybrid Data-Aided Method

Traditional power flow methods often adopt certain assumptions designed for passive balanced distribution systems, thus lacking practicality for unbalanced operation. moreover, their computation accuracy and efficiency are heavily subject to unknown errors and bad data in measurements or prediction data of distributed energy resources (ders). to address these issues, this paper proposes a hybrid data-aided robust power flow algorithm in unbalanced distribution systems, which combines taylor series expansion knowledge with a data-driven regression technique. the proposed method initiates a linearization power flow model to derive an explicitly analytical solution by modified taylor expansion. to mitigate the approximation loss that surges due to the der integration and bad data, we further develop a data-aided robust support vector regression approach to estimate the errors efficiently. comparative analysis in the 13-bus and 123-bus ieee unbalanced feeders shows that the proposed hybrid algorithm achieves superior computational efficiency, with guaranteed accuracy and robustness against outliers.

data-driven↗
DOE OSTI2018

Data-Driven Robust M-LS-SVR-Based NARX Modeling for Estimation and Control of Molten Iron Quality Indices in Blast Furnace Ironmaking

Optimal operation of an industrial blast furnace (BF) ironmaking process largely depends on a reliable measurement of molten iron quality (MIQ), which is not feasible using the conventional sensors. This paper proposes a novel data-driven robust modeling method for online estimation and control of MIQ indices. First, to completely capture the nonlinear dynamics of the BF process, the nonlinear autoregressive exogenous (NARX) model is constructed for the MIQ indices. Then, considering that the standard least-squares support vector regression (LS-SVR) cannot directly cope with the multi-output problem, the multi- task transfer learning is proposed to design a novel multi-output LS-SVR (M-LS-SVR) for learning of the NARX model. Furthermore, a novel M-estimator is proposed to reduce the interference of outliers and improve the robustness of the M-LS- SVR model. Since the weights of different outlier data are properly given by the weight function, their corresponding contributions on modeling can be properly distinguished, thus a robust modeling result can be achieved. Finally, a novel multi- objective evaluation index on modeling performance is developed by comprehensively considering the root mean square error of modeling and the correlation coefficient on trend fitting, based on which the NSGA-II algorithm is used to globally optimize the model parameters. Both experiments using industrial data and industrial applications illustrate that the proposed method can eliminate the adverse effect caused by the fluctuation of data in BF process efficiently, which indicates its stronger robustness and higher accuracy. Moreover, control testing shows that the developed model can be well applied to realize data-driven control of the BF process.

blast cavities, Control↗
DOE OSTI2019

Quantifying the Biophysical and Socioeconomic Drivers of Changes in Forest and Agricultural Land in South and Southeast Asia

South and Southeast Asia (SSEA) has been a hotspot for land use and land cover change (LULCC) in the past few decades. The identification and quantification of the drivers of LULCC are crucial for improving our understanding of LULCC trends. So far, the biophysical and socioeconomic drivers of forest change have not been quantified at the regional scale, particularly for SSEA. In this study, we quantify the biophysical and socioeconomic drivers of forest change on a country-by-country basis in SSEA using an integrated quantitative methodology, which systematically accounts for previously published driver information and regional datasets. We synthesize more than 200 publications to identify the drivers of the forest change at different spatial scales in SSEA. Subsequently, we collect spatially explicit proxy data to represent the identified drivers. We quantify the dynamics of forest and agricultural land from 1992 to 2015 using the Climate Change Initiative (CCI) land cover data developed by the European Space Agency (ESA). A geographically weighted regression method is employed to quantify the spatially heterogeneous drivers of forest change. Our results show that socioeconomic drivers are more important than biophysical drivers for the conversion of forest to agricultural land in South Asia and maritime Southeast Asia. In contrast, biophysical drivers are more important than socioeconomic drivers for the conversion of agricultural land to forest in maritime Southeast Asia and less important in South Asia. Both biophysical and socioeconomic drivers contribute approximately equally to both changes in the mainland Southeast Asia region. By quantifying the dynamics of forest and agricultural land and the spatially explicit drivers of their changes in SSEA, this study provides a solid foundation for LULCC modeling and projection.

Xu, Xiaoming↗
DOE OSTI2022

Initial soil organic carbon stocks govern changes in soil carbon: Reality or artifact?

Abstract Changes in soil organic carbon (SOC) storage have the potential to affect global climate; hence identifying environments with a high capacity to gain or lose SOC is of broad interest. Many cross‐site studies have found that SOC‐poor soils tend to gain or retain carbon more readily than SOC‐rich soils. While this pattern may partly reflect reality, here we argue that it can also be created by a pair of statistical artifacts. First, soils that appear SOC‐poor purely due to random variation will tend to yield more moderate SOC estimates upon resampling and hence will appear to accrue or retain more SOC than SOC‐rich soils. This phenomenon is an example of regression to the mean. Second, normalized metrics of SOC change—such as relative rates and response ratios—will by definition show larger changes in SOC at lower initial SOC levels, even when the absolute change in SOC does not depend on initial SOC. These two artifacts create an exaggerated impression that initial SOC stocks are a major control on SOC dynamics. To address this problem, we recommend applying statistical corrections to eliminate the effect of regression to the mean, and avoiding normalized metrics when testing relationships between SOC change and initial SOC. Careful consideration of these issues in future cross‐site studies will support clearer scientific inference that can better inform environmental management.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2018

Environmental factors function as constraints on soil nitrous oxide fluxes in bioenergy feedstock cropping systems

Abstract Nitrous oxide (N 2 O) is a potent greenhouse gas and major component of the net global warming potential of bioenergy feedstock cropping systems. Numerous environmental factors influence soil N 2 O production, making direct correlation difficult to any one factor of N 2 O fluxes under field conditions. We instead employed quantile regression to evaluate whether soil temperature, water‐filled pore space ( WFPS ), and concentrations of soil nitrate ( ) and ammonium ( ) determined upper bounds for soil N 2 O flux magnitudes. We collected data over 6 years from a range of bioenergy feedstock cropping systems including no‐till grain crops, perennial warm‐season grasses, hybrid poplar, and polycultures of tallgrass prairie species each with and without nitrogen (N) addition grown at two sites. The upper bounds for soil N 2 O fluxes had a significant and positive correlation with all four environmental factors, although relatively large fluxes were still possible at minimal values for nearly all factors. The correlation with was generally weaker, suggesting it is less important than in driving large fluxes. Quantile regression slopes were generally lower for unfertilized perennials than for other systems, but this may have resulted from a perpetual state of nitrogen limitation, which prevented other factors from being clear constraints. This framework suggests efforts to reduce concentrations of in the soil may be effective at reducing high‐intensity periods—”hot moments”—of N 2 O production.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Using Machine Learning Tools to Predict Compressor Stall

Clean energy has become an increasingly important consideration in today’s power systems. As the push for clean energy continues, many coal-fired power plants are being decommissioned in favor of renewable power sources such as wind and solar. However, the intermittent nature of renewables means that dynamic load following traditional power systems is crucial to grid stability. With high flexibility and fast response at a wide range of operating conditions, gas turbine systems are poised to become the main load following component in the power grid. Yet, rapid changes in load can lead to fluid flow instabilities in gas turbine power systems. These instabilities often lead to compressor surge and stall, which are some of the most critical problems facing the safe and efficient operation of compressors in turbomachinery today. Although the topic of compressor surge and stall has been extensively researched, no methods for early prediction have been proven effective. This study explores the utilization of machine learning tools to predict compressor stall. The long short-term memory (LSTM) model, a form of recurrent neural network (RNN), was trained using real compressor stall datasets from a 100 kW recuperated gas turbine power system designed for hybrid configuration. Two variations of the LSTM model, classification and regression, were tested to determine optimal performance. The regression scheme was determined to be the most accurate approach, and a tool for predicting compressor stall was developed using this configuration. Overall, results show that the tool is capable of predicting stalls 5–20 ms before they occur. With a high-speed controller capable of 5 ms time-steps, mitigating action could be taken to prevent compressor stall before it occurs.

42 ENGINEERING↗
DOE OSTI2014

Evaluation of image guided motion management methods in lung cancer radiotherapy

Purpose: To evaluate the accuracy and reliability of three target localization methods for image guided motion management in lung cancer radiotherapy. Methods: Three online image localization methods, including (1) 2D method based on 2D cone beam (CB) projection images, (2) 3D method using 3D cone beam CT (CBCT) imaging, and (3) 4D method using 4D CBCT imaging, have been evaluated using a moving phantom controlled by (a) 1D theoretical breathing motion curves and (b) 3D target motion patterns obtained from daily treatment of 3 lung cancer patients. While all methods are able to provide target mean position (MP), the 2D and 4D methods can also provide target motion standard deviation (SD) and excursion (EX). For each method, the detected MP/SD/EX values are compared to the analytically calculated actual values to calculate the errors. The MP errors are compared among three methods and the SD/EX errors are compared between the 2D and 4D methods. In the theoretical motion study (a), the dependency of MP/SD/EX error on EX is investigated with EX varying from 2.0 cm to 3.0 cm with an increment step of 0.2 cm. In the patient motion study (b), the dependency of MP error on target sizes (2.0 cm and 3.0 cm), motion patterns (four motions per patient) and EX variations is investigated using multivariant linear regression analysis. Results: In the theoretical motion study (a), the MP detection errors are −0.2 ± 0.2, −1.5 ± 1.1, and −0.2 ± 0.2 mm for 2D, 3D, and 4D methods, respectively. Both the 2D and 4D methods could accurately detect motion pattern EX (error < 1.2 mm) and SD (error < 1.0 mm). In the patient motion study (b), MP detection error vector (mm) with the 2D method (0.7 ± 0.4) is found to be significantly less than with the 3D method (1.7 ± 0.8,p < 0.001) and the 4D method (1.4 ± 1.0, p < 0.001) using paired t-test. However, no significant difference is found between the 4D method and the 3D method. Based on multivariant linear regression analysis, the variances of MP error in SI direction explained by target sizes, motion patterns, and EX variations are 9% with the 2D method, 74.4% with the 3D method, and 27% with the 4D method. The EX/SD detection errors are both < 1.0 mm for the 2D method and < 2.0 mm for the 4D method. Conclusions: The 2D method provides the most accurate MP detection regardless of the motion pattern variations, while its performance is limited by the accuracy of target identification in the projection images. The 3D method causes the largest error in MP determination, and its accuracy significantly depends on target sizes, motion patterns, and EX variations. The 4D method provides moderate MP detection results, while its accuracy relies on a regular motion pattern. In addition, the 2D and 4D methods both provide accurate measurement of the motion SD/EX, providing extra information for motion management.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2014

SU-E-J-237: Image Feature Based DRR and Portal Image Registration

Purpose: Two-dimensional (2D) matching of the kV X-ray and digitally reconstructed radiography (DRR) images is an important setup technique for image-guided radiotherapy (IGRT). In our clinics, mutual information based methods are used for this purpose on commercial linear accelerators, but with often needs for manual corrections. This work proved the feasibility that feature based image transform can be used to register kV and DRR images. Methods: The scale invariant feature transform (SIFT) method was implemented to detect the matching image details (or key points) between the kV and DRR images. These key points represent high image intensity gradients, and thus the scale invariant features. Due to the poor image contrast from our kV image, direct application of the SIFT method yielded many detection errors. To assist the finding of key points, the center coordinates of the kV and DRR images were read from the DICOM header, and the two groups of key points with similar relative positions to their corresponding centers were paired up. Using these points, a rigid transform (with scaling, horizontal and vertical shifts) was estimated. We also artificially introduced vertical and horizontal shifts to test the accuracy of our registration method on anterior-posterior (AP) and lateral pelvic images. Results: The results provided a satisfactory overlay of the transformed kV onto the DRR image. The introduced vs. detected shifts were fit into a linear regression. In the AP image experiments, linear regression analysis showed a slope of 1.15 and 0.98 with an R2 of 0.89 and 0.99 for the horizontal and vertical shifts, respectively. The results are 1.2 and 1.3 with R2 of 0.72 and 0.82 for the lateral image shifts. Conclusion: This work provided an alternative technique for kV to DRR alignment. Further improvements in the estimation accuracy and image contrast tolerance are underway.

62 RADIOLOGY AND NUCLEAR MEDICINE↗