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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
NASA NTRS2015

The Use of Absolute-Value Terms in Regression Modeling of Multi-Piece Force Balances

Different aspects of the use of absolute-value terms in regression models of the electrical outputs of multi-piece force balance calibration data are discussed. First, characteristics of a variety of regression model term combinations with absolute-value terms are reviewed that are currently used in the aerospace testing community to fit the gage outputs of a balance. Then, a semi-empirical test is presented that quantifies bidirectional characteristics of the balance bridge outputs. Several diagnostic methods are discussed to assess the severity of near-linear dependencies between regressors of models with absolute-value terms. In particular, connections between the linear, absolute-value, quadratic, signed quadratic, and cubic terms are studied in greater detail. Data from an automated calibration of NASAs MK29B force balance are used to illustrate the most important observations and results. Rules of thumb that variance-inflation factors be less than 10 must be relaxed when using absolute-value terms to describe bidirectional balances.

calibration analysis↗
NASA NTRS2019

Winter Wheat Yield Assessment from Landsat 8 and Sentinel-2 Data: Incorporating Surface Reflectance, Through Phenological Fitting, into Regression Yield Models

A combination of Landsat 8 and Sentinel-2 offers a high frequency of observations (3–5 days) at moderate spatial resolution (10–30 m), which is essential for crop yield studies. Existing methods traditionally apply vegetation indices (VIs) that incorporate surface reflectances (SRs) in two or more spectral bands into a single variable, and rarely address the incorporation of SRs into empirical regression models of crop yield. In this work, we address these issues by normalizing satellite data (both VIs and SRs) derived from NASA’s Harmonized Landsat Sentinel-2 (HLS) product, through a phenological fitting. We apply a quadratic function to fit VIs or SRs against accumulated growing degree days (AGDDs), which affects the rate of crop development. The derived phenological metrics for VIs and SRs, namely peak, area under curve (AUC), and fitting coefficients from a quadratic function, were used to build empirical regression winter wheat models at a regional scale in Ukraine for three years, 2016–2018. The best results were achieved for the model with near infrared (NIR) and red spectral bands and derived AUC, constant, linear, and quadratic coefficients of the quadratic model. The best model yielded a root mean square error (RMSE) of 0.201 t/ha (5.4%) and coefficient of determination R2 = 0.73 on cross-validation.

phenological fitting↗
NASA NTRSDate not supplied

Bingo: A Customizable Framework for Symbolic Regression with Genetic Programming

In this paper, we introduce Bingo, a flexible and customizable yet performant Python framework for symbolic regression with genetic programming. Bingo maintains a modular code structure for simple abstraction and easily swappable components. Fitness functions, selection methods, and constant optimization methods allow for easy problem-specific customization. Bingo also maintains several features for increased efficiency such as parallelism, equation simplification, and a C++ backend. We compare Bingo’s performance to other genetic programming for symbolic regression (GPSR) methods to show that it is both competitive and flexible.

David Randall↗
NASA NTRS2022

Updated Trends of the Stratospheric Ozone Vertical Distribution in the 60°S-60°N Latitude Range Based on the LOTUS Regression Model

This study presents an updated evaluation of stratospheric ozone profile trends in the 60°S - 60°N latitude range over the 2000 - 2020 period using an updated version of the Long-term Ozone Trends and Uncertainties in the Stratosphere (LOTUS) regression model that was used to evaluate such trends up to 2016 for the last WMO Ozone Assessment (2018). In addition to the derivation of detailed trends as a function of latitude and vertical coordinates, the regressions are performed with the data sets averaged over broad latitude bands, i.e., 60°S–35°S, 20°S–20°N and 35°N–60°N. The same methodology as in the last Assessment is applied to combine trends in these broad latitude bands in order to compare the results with the previous studies. Longitudinally resolved merged satellite records are also considered in order to provide a better comparison with trends retrieved from ground-based records, e.g., lidar, ozone sondes, Umkehr, microwave and Fourier Transform Infrared (FTIR) spectrometers at selected stations where long-term time series are available. The study includes a comparison with trends derived from the REF-C2 simulations of the Chemistry Climate Model Initiative (CCMI-1). This work confirms past results showing an ozone increase in the upper stratosphere, which is now significant in the three broad latitude bands. The increase is largest in the northern and southern hemisphere midlatitudes, with ~2.2%/decade at ~2.1 hPa, and ~2.1%/decade at ~3.2 hPa respectively, compared to ~1.6%/decade at ~2.6 hPa in the tropics. New trend signals have emerged from the records, such as a significant decrease of ozone in the tropics around 35 hPa and a non-significant increase of ozone in the southern midlatitudes at about 20 hPa. Non-significant negative ozone trends are derived in the lowermost stratosphere, with the most pronounced trends in the tropics. While a very good agreement is obtained between trends from merged satellite records and the CCMI-1 REF-C2 simulation in the upper stratosphere, observed negative trends in the lower stratosphere are not reproduced by models at southern and, in particular, at northern midlatitudes, where models report an ozone increase. However, the lower stratospheric trend uncertainties are quite large, for both measured and modelled trends. Finally, 2000-2020 stratospheric ozone trends derived from the ground-based and longitudinally resolved satellite records are in reasonable agreement over the European Alpine and tropical regions, while at the Lauder station in the southern hemisphere mid-latitudes they show some differences.

Stratospheric ozone trends↗
NASA NTRSDate not supplied

Component Level Regression Testing in a Hierarchical Architecture

The Goddard Earth Observing System (GEOS) is an Earth system model consisting of a large suite of individual model components that can be coupled in a flexible manner to investigate a variety of Earth science issues. Specific GEOS model configurations are composed as a hierarchical collection of components based on the Earth System Modeling Framework (ESMF). Regression testing of GEOS is currently limited to (1) full system tests that are poor at isolating specific defects and (2) a suite of unit tests which have very limited coverage. As part of our approach to improve upon the current testing situation, we have prototyped the capability to perform regression tests on individual GEOS components by leveraging and extending existing checkpoint/restart capabilities. In our implementation, each ESMF component has 3 states: Import (what it needs to run), Export (which it needs to provide to other components), and Internal (the component state proper). By capturing, Import, Export and Internal states for a given component during a ull run of GEOS, a generic driver can then rerun the component offline and compare expected exports with those that have been saved. The hierarchical structure of GEOS introduces an interesting wrinkle when trying to test components that in turn drive interacting child components. To fully isolate a parent component, we use the approach of software mocks, in which the exports of children are also saved during the initial capture run of GEOS. Then when testing the parent component, the children components are replaced by a generic mock component that produces exports from the previously saved data and ensures that that all interdependencies among children components are satisfied.

Thomas Clune↗
DOE OSTI2016

The Value of Regression Models in Determining Industrial Energy Savings

Determination of industrial facility energy savings is commonly undertaken to comply with legal requirements, meet sustainability goals, and quantify greenhouse gas emissions, cost savings, and as one way to demonstrate energy performance improvement per the requirements of an energy management system standard, such as ISO 50001. At their most basic level, energy savings are determined through a process of accounting for baseline and reporting period energy consumption followed by a calculation of energy savings as the difference in the levels of energy consumption between the two periods. The determination of industrial facility-wide energy savings can be undertaken in a wide variety of ways and in relation to energy performance improvement, energy savings can be expressed as any number of energy performance indicators (EnPIs). Energy savings values will vary depending upon if an absolute, intensity, or regression basis of calculation is selected for use. The use of one of the three basis of energy savings calculation will alter the relevance, meaning, and comparability of the resulting energy savings values To increase consistency in the determination of energy savings values, standards such as the International Performance Measurement and Verification Protocol (IPMVP), ISO 17747 – Determination of Energy Savings in Organization, and the U.S. Department of Energy (U.S. DOE) Superior Energy Performance (SEP) program Measurement and Verification Protocol (SEP M&V Protocol) have been published. Industrial facilities in the United States, Mexico, and Canada are certified to the U.S. DOE SEP program after becoming ISO 50001 certified and verification by a third party of calculated energy savings per the SEP M&V Protocol. This paper discusses some of the different meanings energy savings values will have when calculated by using the three different bases. Additionally, data from five SEP certified industrial facilities are used to calculate energy savings values using the three bases and compared as energy performance improvement percentage (EPI %) values. Discussion of why a large degree in variation in EPI % values are seen for any given set of data is provided. Ultimately, while each energy saving calculation basis has merit and can be used appropriately in its own context, the regression basis is shown to best translates energy savings values into contextualized energy performance improvement values by conveying comparable information regarding operational and behavioural improvements not otherwise observable with the absolute or intensity bases.

Therkelsen, Peter↗
DOE OSTI2024

Multiobjective Constrained Symbolic Regression for Predictive Modeling of Material Creep Behavior

When creep testing is repeated on samples of the same alloy under the same parametric conditions (i.e., stress and temperature), the resulting strain/time curves can vary from each other considerably as shown in Figure 1 [1]. The time required to creep test a material to rupture can extend to the order of years. Because of this, a numerical model that can quickly analyze the incomplete results of an ongoing experiment to predict 1) the incomplete portion of the strain/time curve leading up to the rupture point and 2) the rupture point itself would be of great utility to the materials community. Such a model has the potential to save 1) the time required to finish running the experiment to rupture 2) the associated monetary cost of finishing said experiment. Furthermore, it would be advantageous if the predictive model could give a parametric function modeling strain/time curves for material scientists to investigate the impact of the temperature and stress parameters on the resulting creep behavior. This work introduces a piecewise symbolic regression algorithm to predict the remainder of the strain/time curve. Preliminary results show good model performance.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Proxy quality control of biomass particles using thermogravimetric analysis and Gaussian process regression models

Abstract The temperature experienced by reactants during preparation in a reactor is a key component in determining the yield and homogeneity of usable chemical products such as biomass particles. Thermocouples with sensors can be used to monitor spatial temperature gradients within reactors but these sensors are often too expensive and/or invasive. The present work proposes a strategy to identify optimal machine learning models to infer the maximum effective temperature experienced by particles during oxidative biomass torrefaction using key thermochemical combustion parameters. The maximum rate of weight loss, the corresponding temperature, and fixed carbon content on a dry‐ash‐free basis are used as literature‐based predictor variables obtained from thermogravimetric analysis. The evaluation of 24 machine‐learning models using the standard tenfold cross‐validation method suggests that the exponential Gaussian process regression (GPR) model is the most effective, followed by other GPR models. These high‐performing GPR models were also utilized to predict the effective preparation temperature distribution of reactor‐produced biomass particles under eight conditions of varying residence time and air‐to‐biomass ratio. The effective preparation temperature and residence time of individual biomass particles were then encoded into the torrefaction severity factor and used to estimate the energy yield of the reactor output as a novel quality control method. © 2023 The Authors. Biofuels, Bioproducts and Biorefining published by Society of Industrial Chemistry and John Wiley & Sons Ltd.

09 BIOMASS FUELS↗
DOE OSTI2023

Probabilistic partition of unity networks for high–dimensional regression problems

We explore the probabilistic partition of unity network (PPOU-Net) model in the context of high-dimensional regression problems and propose a general framework focusing on adaptive dimensionality reduction. With the proposed framework, the target function is approximated by a mixture of experts model on a low-dimensional manifold, where each cluster is associated with a fixed-degree polynomial. We present a training strategy that leverages the expectation maximization (EM) algorithm. During the training, we alternate between (i) applying gradient descent to update the DNN coefficients; and (ii) using closed-form formulae derived from the EM algorithm to update the mixture of experts model parameters. Under the probabilistic formulation, step (ii) admits the form of embarrassingly paralleliazable weighted least-squares solves. The PPOU-Nets consistently outperform the baseline fully-connected neural networks of comparable sizes in numerical experiments of various data dimensions. Here, we also explore the proposed model in applications of quantum computing, where the PPOU-Nets act as surrogate models for cost landscapes associated with variational quantum circuits.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Linear Regression Based Multi-fidelity Surrogate for Disturbance Amplification in Multi-phase Explosion

When simulations are very expensive and many are required, as for optimization or uncertainty quantification, a way to reduce cost is using surrogates. With multiple simulations to predict the quantity of interest, some being very expensive and accurate (high-fidelity simulations) and others cheaper but less accurate (low-fidelity simulations), it may be worthwhile to use multifidelity surrogates (MFSs). Moreover, if we can afford just a few high-fidelity simulations or experiments, MFS becomes necessary. Co-Kriging, which is probably the most popular MFS, replaces both low-fidelity and high-fidelity simulations by a single MFS. A recently proposed linear regression–based MFS (LR-MFS) offers the option to correct the LF simulations instead of correcting the LF surrogate in the MFS. When the low-fidelity simulation is cheap enough for use in an application, such as optimization, this may be an attractive option. Here, we explore the performance of LR-MFS using exact and surrogate-replaced low-fidelity simulations. The concern studied is a cylindrical dispersal of 100-μ m-diameter solid particles after detonation and the quantity of interest is a measure of the amplification of the departure from axisymmetry. We find very substantial accuracy improvements for this problem using the LR-MFS with exact low-fidelity simulations. Inspired by these results, we also compare the performance of co-Kriging to the use of Kriging to correct exact low-fidelity simulations and find a similar accuracy improvement when simulations are directly used. For this problem, further improvements in accuracy are achievable by taking advantage of inherent parametric symmetries. These findings may alert users of MFSs to the possible advantages of using exact low-fidelity simulations when this is affordable.

42 ENGINEERING↗
DOE OSTI2018

Two-Step Estimation in a Heteroscedastic Linear Regression Model

We study the problem of estimating a parameter in some heteroscedastic linear regression model in the case where the regressors consist of all order statistics based on the sample of identically distributed not necessarily independent observations with finite second moment. It is assumed that the random errors depend on the parameter and distributions of the corresponding regressors. We propose a two-step procedure for finding explicit asymptotically normal estimators.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Prediction of Creep-Induced Strain Using a Symbolic Regression-Based Model

Material creep under high-temperature conditions limits the lifetime and safety of structural systems such as advanced nuclear reactors. Conventional creep testing is slow and often produces inconsistent results across nominally identical experiments, making lifetime prediction uncertain. Here, to address these challenges, this work develops a data-driven symbolic regression (SR) model that consolidates results from duplicate creep tests and predicts the remaining strain-time curve of an ongoing experiment. The method uses piece-wise multi-objective SR with physical constraints to generate analytic, interpretable functions describing transient creep strain. Applied to Inconel Alloy 617 data, the approach achieved relative mean absolute errors of 1.0–9.5%, providing closed-form predictions of strain evolution. These results demonstrate a first step toward reducing the duration and cost of long-term creep testing while retaining physically interpretable model forms.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2018

A new accelerated algorithm for ill-conditioned ridge regression problems

In this work, we study the primal and dual formulations of the regularized least squares problem, in special norm L{sub 2}, named ridge regression. We state the Gradient method for the corresponding primal-dual problem through a fixed point approach. For this formulation, we present an original convergence analysis involving the spectral radius of a suitable iteration matrix. The main contribution of this work is to accelerate the studied method by means of a memory-like strategy. Some connections with the spectral properties of the associated iteration matrices were established to prove the convergence of the accelerated method. Preliminary experiments showed the good performance of the proposed algorithm when applied to solve ill-conditioned problems, providing better results than the conjugate gradient method for this class of problems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Data-driven surrogates for high dimensional models using Gaussian process regression on the Grassmann manifold

This paper introduces a surrogate modeling scheme based on Grassmannian manifold learning to be used for cost-efficient predictions of high-dimensional stochastic systems. The method exploits subspace-structured features of each solution by projecting it onto a Grassmann manifold. This point-wise linear dimensionality reduction harnesses the structural information to assess the similarity between solutions at different points in the input parameter space. The method utilizes a solution clustering approach in order to identify regions of the parameter space over which solutions are sufficiently similarly such that they can be interpolated on the Grassmannian. In this clustering, the reduced-order solutions are partitioned into disjoint clusters on the Grassmann manifold using the eigen-structure of properly defined Grassmannian kernels and, the Karcher mean of each cluster is estimated. Then, the points in each cluster are projected onto the tangent space with origin at the corresponding Karcher mean using the exponential mapping. For each cluster, a Gaussian process regression model is trained that maps the input parameters of the system to the reduced solution points of the corresponding cluster projected onto the tangent space. Using this Gaussian process model, the full-field solution can be efficiently predicted at any new point in the parameter space. In certain cases, the solution clusters will span disjoint regions of the parameter space. In such cases, for each of the solution clusters we utilize a second, density-based spatial clustering to group their corresponding input parameter points in the Euclidean space. The proposed method is applied to two numerical examples. Here, the first is a nonlinear stochastic ordinary differential equation with uncertain initial conditions where the surrogate is used to predict the time history solution. The second involves modeling of plastic deformation in a model amorphous solid using the Shear Transformation Zone theory of plasticity, where the proposed surrogate is used to predict the full strain field of a material specimen under large shear strains.

42 ENGINEERING↗
DOE OSTI2024

A crystal-plasticity-informed Gaussian Process Regression model to capture anisotropy in single crystal shape memory alloys

This work presents a machine learning (ML) framework that model the anisotropic actuation responses in a shape memory alloy. A Gaussian Process Regression (GPR) based ML model is trained on a set of different crystal orientations subjected to different actuation conditions. The training employed thermo-mechanical responses from a crystal-plasticity model that captures phase-transformation, stress-induced plasticity, and transformation-induced plasticity. Further, on training the GPR-ML model at fixed stress level for different orientations, it captured the thermo-mechanical responses accounting for the anisotropy, and predicted responses for new orientations with good accuracy. The GPR-ML model is able to capture the transformation temperature variations even when trained using multiple stress levels, and the transformation strain showed significant deviations. The developed GPR-ML model gave reasonable predictions for an unexplored sample set of orientations and loading conditions.

36 MATERIALS SCIENCE↗
DOE OSTI2018

Ranking the importance of Wildfires' human drivers through a multi-model regression approach

Highlights: • Spatial variability of wildfire regimes is closely related to human activities. • Fire variables have different correlation structures with socioeconomic variables. • Rural poverty, economic disparities and social conflicts are associated to high fire risk. • Landscape structure, land-use and cropping systems influence local-scale fire regimes. • Local unemployment rates discriminate among different wildfire regimes. - Abstract: Wildfires are a major ecological disturbance in Mediterranean environments, and affect together natural resources, ecosystem services and human activities. The impact of socioeconomic forces on wildfire regimes is generally less investigated than the effects of biophysical drivers. Being grounded on a multi-model regression analysis of socioeconomic and territorial indicators, the present study identifies relevant factors influencing local-scale wildfires' regimes in Italy. An economically-disadvantaged context with persistent unemployment, rural poverty, social inequalities and population aging proved to be associated to more frequent fire events. Additionally, our analysis points out that occurrence, average size and density of wildfires reflect different correlation profiles with the local-scale socioeconomic context. Along with the socioeconomic profile of local communities, the empirical outcomes of our study show the importance of landscape structure, land-use and cropping systems in local-scale fire regimes. The empirical results of this study justify a multidimensional analysis of relevant socioeconomic dimensions in fire risk assessment and contribute to an informed approach to wildfire management. Moreover, our study provides basic knowledge advancing research on fire prevention, and informing spatial planning and developmental policies aimed at increasing preparedness to large fires.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Uncertain characterization of reservoir fluids due to brittleness of equation of state regression

Equations of state (EoS) play a central role in modeling the phase equilibrium of fluid mixtures. Their parameterization involves fitting a model to experimental data, i.e., solving a nonlinear, non-convex, multivariate optimization problem. The latter requires one to select design variables, domains of definition for each variable, and weights assigned to individual measurements. We demonstrate that subjective choices of an optimization algorithm and an initial guess also impact the regression process. Consequently, EoS predictions are fundamentally uncertain even after the EoS tuning to a limited set of experimental data points. We demonstrate this observation for two hydrocarbon reservoir fluids, in which five properties of the heaviest carbon fraction are treated as design variables. While all the optimization algorithms and initial guesses match experimental data for the gas and liquid properties, the resulting EoS parameterizations lead to dramatically different predictions of the fluid’s thermophysical behavior in the unsampled pressure and temperature regions. In conclusion, we propose the probabilistic treatment of design variables to quantify the predictive uncertainty of the resulting fluid models.

15 GEOTHERMAL ENERGY↗
DOE OSTI2026

Efficient data-driven regression for reduced-order modeling of spatial pattern formation

We present an efficient data-driven regression approach for constructing reduced-order models (ROMs) of reaction-diffusion systems exhibiting pattern formation. The ROMs are learned non-intrusively from available training data of physically accurate numerical simulations. The method can be applied to general nonlinear systems through the use of polynomial model form, while not requiring knowledge of the underlying physical model, governing equations, or numerical solvers. The process of learning ROMs is posed as a low-cost least-squares problem in a reduced-order subspace identified via Proper Orthogonal Decomposition (POD). Numerical experiments on classical pattern-forming systems–including the Schnakenberg and Mimura–Tsujikawa models–demonstrate that higher-order surrogate models significantly improve prediction accuracy while maintaining low computational cost. The proposed method provides a flexible, non-intrusive model reduction framework, well suited for the analysis of complex spatio-temporal pattern formation phenomena.

Data-driven modeling↗