Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “nonlinear 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 325 records · Page 18
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Adaptive Online Multivariate Signal Extraction With Locally Weighted Robust Polynomial Regression

High-frequency, multivariate data collected in real-time and used to control or make decisions regarding a process’ operation often contain some noise and outliers. Thus, a method to extract the signal is needed in order to reduce the number and magnitude of control-based adjustments that are implemented. Such a method must be (i) online, depending only on past and current observations; (ii) fast, producing a smooth value more quickly than the measurement frequency; (iii) robust, ignoring brief bursts of erroneously measured values; (iv) multivariate, ignoring observations that are jointly unusual; (v) adaptive, adjusting to periods of rapid fluctuation in the signal versus periods of stability; and (vi) purely data-driven, not incorporating any information about the process from which the data are collected. Most existing methods are only able to address a subset of these six features. Furthermore, we also require the method to be nonlinear, providing a local nonlinear estimate of the signal. In this work, we propose a novel, real-time signal extraction method based on a local, robust polynomial fit. We demonstrate the performance of our method compared to a state-of-the-art competitor through simulation. For illustration, the methodology is applied to data collected from a reverse osmosis water treatment process.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2007

Twist Model Development and Results from the Active Aeroelastic Wing F/A-18 Aircraft

Understanding the wing twist of the active aeroelastic wing (AAW) F/A-18 aircraft is a fundamental research objective for the program and offers numerous benefits. In order to clearly understand the wing flexibility characteristics, a model was created to predict real-time wing twist. A reliable twist model allows the prediction of twist for flight simulation, provides insight into aircraft performance uncertainties, and assists with computational fluid dynamic and aeroelastic issues. The left wing of the aircraft was heavily instrumented during the first phase of the active aeroelastic wing program allowing deflection data collection. Traditional data processing steps were taken to reduce flight data, and twist predictions were made using linear regression techniques. The model predictions determined a consistent linear relationship between the measured twist and aircraft parameters, such as surface positions and aircraft state variables. Error in the original model was reduced in some cases by using a dynamic pressure-based assumption. This technique produced excellent predictions for flight between the standard test points and accounted for nonlinearities in the data. This report discusses data processing techniques and twist prediction validation, and provides illustrative and quantitative results.

Lizotte, Andrew M.↗
DOE OSTI2023

Machine learning reduces soft costs for residential solar photovoltaics

Further deployment of rooftop solar photovoltaics (PV) hinges on the reduction of soft (non-hardware) costs—now larger and more resistant to reductions than hardware costs. The largest portion of these soft costs is the expenses solar companies incur to acquire new customers. In this study, we demonstrate the value of a shift from significance-based methodologies to prediction-oriented models to better identify PV adopters and reduce soft costs. We employ machine learning to predict PV adopters and non-adopters, and compare its prediction performance with logistic regression, the dominant significance-based method in technology adoption studies. Our results show that machine learning substantially enhances adoption prediction performance: The true positive rate of predicting adopters increased from 66 to 87%, and the true negative rate of predicting non-adopters increased from 75 to 88%. We attribute the enhanced performance to complex variable interactions and nonlinear effects incorporated by machine learning. With more accurate predictions, machine learning is able to reduce customer acquisition costs by 15% ($0.07/Watt) and identify new market opportunities for solar companies to expand and diversify their customer bases. Our research methods and findings provide broader implications for the adoption of similar clean energy technologies and related policy challenges such as market growth and energy inequality.

14 SOLAR ENERGY↗
DOE OSTI2023

Direct Estimation of Parameters in ODE Models Using WENDy: Weak-Form Estimation of Nonlinear Dynamics

Abstract We introduce the Weak-form Estimation of Nonlinear Dynamics (WENDy) method for estimating model parameters for non-linear systems of ODEs. Without relying on any numerical differential equation solvers, WENDy computes accurate estimates and is robust to large (biologically relevant) levels of measurement noise. For low dimensional systems with modest amounts of data, WENDy is competitive with conventional forward solver-based nonlinear least squares methods in terms of speed and accuracy. For both higher dimensional systems and stiff systems, WENDy is typically both faster (often by orders of magnitude) and more accurate than forward solver-based approaches. The core mathematical idea involves an efficient conversion of the strong form representation of a model to its weak form, and then solving a regression problem to perform parameter inference. The core statistical idea rests on the Errors-In-Variables framework, which necessitates the use of the iteratively reweighted least squares algorithm. Further improvements are obtained by using orthonormal test functions, created from a set of $$C^{\infty }$$ C ∞ bump functions of varying support sizes.We demonstrate the high robustness and computational efficiency by applying WENDy to estimate parameters in some common models from population biology, neuroscience, and biochemistry, including logistic growth, Lotka-Volterra, FitzHugh-Nagumo, Hindmarsh-Rose, and a Protein Transduction Benchmark model. Software and code for reproducing the examples is available at https://github.com/MathBioCU/WENDy .

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1998

Experiments to Determine Whether Recursive Partitioning (CART) or an Artificial Neural Network Overcomes Theoretical Limitations of Cox Proportional Hazards Regression

New computationally intensive tools for medical survival analyses include recursive partitioning (also called CART) and artificial neural networks. A challenge that remains is to better understand the behavior of these techniques in effort to know when they will be effective tools. Theoretically they may overcome limitations of the traditional multivariable survival technique, the Cox proportional hazards regression model. Experiments were designed to test whether the new tools would, in practice, overcome these limitations. Two datasets in which theory suggests CART and the neural network should outperform the Cox model were selected. The first was a published leukemia dataset manipulated to have a strong interaction that CART should detect. The second was a published cirrhosis dataset with pronounced nonlinear effects that a neural network should fit. Repeated sampling of 50 training and testing subsets was applied to each technique. The concordance index C was calculated as a measure of predictive accuracy by each technique on the testing dataset. In the interaction dataset, CART outperformed Cox (P less than 0.05) with a C improvement of 0.1 (95% Cl, 0.08 to 0.12). In the nonlinear dataset, the neural network outperformed the Cox model (P less than 0.05), but by a very slight amount (0.015). As predicted by theory, CART and the neural network were able to overcome limitations of the Cox model. Experiments like these are important to increase our understanding of when one of these new techniques will outperform the standard Cox model. Further research is necessary to predict which technique will do best a priori and to assess the magnitude of superiority.

Kattan, Michael W.↗
DOE OSTI2015

Cytogenetic Low-Dose Hyperradiosensitivity Is Observed in Human Peripheral Blood Lymphocytes

Purpose: The shape of the ionizing radiation response curve at very low doses has been the subject of considerable debate. Linear-no-threshold (LNT) models are widely used to estimate risks associated with low-dose exposures. However, the low-dose hyperradiosensitivity (HRS) phenomenon, in which cells are especially sensitive at low doses but then show increased radioresistance at higher doses, provides evidence of nonlinearity in the low-dose region. HRS is more prominent in the G2 phase of the cell cycle than in the G0/G1 or S phases. Here we provide the first cytogenetic mechanistic evidence of low-dose HRS in human peripheral blood lymphocytes using structural chromosomal aberrations. Methods and Materials: Human peripheral blood lymphocytes from 2 normal healthy female donors were acutely exposed to cobalt 60 γ rays in either G0 or G2 using closely spaced doses ranging from 0 to 1.5 Gy. Structural chromosomal aberrations were enumerated, and the slopes of the regression lines at low doses (0-0.4 Gy) were compared with doses of 0.5 Gy and above. Results: HRS was clearly evident in both donors for cells irradiated in G2. No HRS was observed in cells irradiated in G0. The radiation effect per unit dose was 2.5- to 3.5-fold higher for doses ≤0.4 Gy than for doses >0.5 Gy. Conclusions: These data provide the first cytogenetic evidence for the existence of HRS in human cells irradiated in G2 and suggest that LNT models may not always be optimal for making radiation risk assessments at low doses.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
NASA NTRS2005

Twist Model Development and Results From the Active Aeroelastic Wing F/A-18 Aircraft

Understanding the wing twist of the active aeroelastic wing F/A-18 aircraft is a fundamental research objective for the program and offers numerous benefits. In order to clearly understand the wing flexibility characteristics, a model was created to predict real-time wing twist. A reliable twist model allows the prediction of twist for flight simulation, provides insight into aircraft performance uncertainties, and assists with computational fluid dynamic and aeroelastic issues. The left wing of the aircraft was heavily instrumented during the first phase of the active aeroelastic wing program allowing deflection data collection. Traditional data processing steps were taken to reduce flight data, and twist predictions were made using linear regression techniques. The model predictions determined a consistent linear relationship between the measured twist and aircraft parameters, such as surface positions and aircraft state variables. Error in the original model was reduced in some cases by using a dynamic pressure-based assumption and by using neural networks. These techniques produced excellent predictions for flight between the standard test points and accounted for nonlinearities in the data. This report discusses data processing techniques and twist prediction validation, and provides illustrative and quantitative results.

Lizotte, Andrew↗
DOE OSTI2022

Demonstration of Decentralized Physics-Driven Learning

In typical artificial neural networks, neurons adjust according to global calculations of a central processor, but in the brain, neurons and synapses self-adjust based on local information. Contrastive learning algorithms have recently been proposed to train physical systems, such as fluidic, mechanical, or electrical networks, to perform machine-learning tasks from local evolution rules. However, to date, such systems have only been implemented in silico due to the engineering challenge of creating elements that autonomously evolve based on their own response to two sets of global boundary conditions. Here, we introduce and implement a physics-driven contrastive learning scheme for a network of variable resistors, using circuitry to locally compare the response of two identical networks subjected to the two different sets of boundary conditions. Using this method, our system effectively trains itself, optimizing its resistance values without the use of a central processor or external information storage. Once the system is trained for a specified allostery, regression, or classification task, the task is subsequently performed rapidly and automatically by the physical imperative to minimize power dissipation in response to the given voltage inputs. Here, we demonstrate that, unlike typical computers, such learning systems are robust to extreme damage (and thus manufacturing defects) due to their decentralized learning. Our twin-network approach is therefore readily scalable to extremely large or nonlinear networks, where its distributed nature will be an enormous advantage; a laboratory network of only 500 edges will already outpace its in silico counterpart.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2013

The relationship between cadmium in kidney and cadmium in urine and blood in an environmentally exposed population

Introduction: Cadmium (Cd) is toxic to the kidney and a major part of the body burden occurs here. Cd in urine (U-Cd) and blood (B-Cd) are widely-used biomarkers for assessing Cd exposure or body burden. However, empirical general population data on the relationship between Cd in kidney (K-Cd), urine, and blood are scarce. Our objectives were to determine the relationship between cadmium in kidney, urine, and blood, and calculate the elimination half-time of Cd from the kidney. Methods: Kidney cortex biopsies, urine, and blood samples were collected from 109 living kidney donors. Cd concentrations were determined and the relationships between K-Cd, U-Cd, and B-Cd were investigated in regression models. The half-time of K-Cd was estimated from the elimination constant. Results: There was a strong association between K-Cd and U-Cd adjusted for creatinine (r{sub p} = 0.70, p < 0.001), while the association with B-Cd was weaker (r{sub p} = 0.44, p < 0.001). The relationship between K-Cd and U-Cd was nonlinear, with slower elimination of Cd at high K-Cd. Estimates of the K-Cd half-time varied between 18 and 44 years. A K-Cd of 25 μg/g corresponds to U-Cd of 0.42 μg/g creatinine in overnight urine (U-Cd/K-Cd ratio: about 1:60). Multivariate models showed Cd in blood and urinary albumin as determinants for U-Cd excretion. Discussion: In healthy individuals with low-level Cd exposure, there was a strong correlation between Cd in kidney and urine, especially after adjustment for creatinine. Urinary Cd was also affected by Cd in blood and urinary albumin. Previous estimates of the U-Cd/K-Cd ratio may underestimate K-Cd at low U-Cd. - Highlights: ► The first study of the relation between Cd in kidney, blood and urine at low U-Cd ► Simultaneous samples were collected from healthy kidney donors. ► There was a nonlinear relationship between cadmium in kidney and urine. ► Estimates of the kidney cadmium half-time were 18–44 years, depending on model used. ► Previous data seem to underestimate kidney cadmium at low urinary cadmium.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2019

Thermal modulation of nonlinear ultrasonic wave for concrete damage evaluation

A nonlinear ultrasonic test is proposed for material damage evaluation using thermal modulation. Temperature change excites and modulates nonlinear behavior of ultrasonic waves in concrete. Coda wave interferometry was used to analyze the relative velocity change of ultrasonic wave with temperature variations. On concrete samples with different levels of thermal damage, experimental results indicate that the samples with a higher damage level demonstrated higher sensitivity to temperature variations. In addition, a slow dynamics nonlinear behavior was observed. When the temperature changed and then remained constant, the wave velocity gradually approached to its equilibrium value.

Acoustic spectroscopy↗
NASA NTRS2012

Cascade Storage and Delivery System for a Multi Mission Space Exploration Vehicle (MMSEV)

NASA is developing a Multi Mission Space Exploration Vehicle (MMSEV) for missions beyond Low Earth Orbit (LEO). The MMSEV is a pressurized vehicle used to extend the human exploration envelope for Lunar, Near Earth Object (NEO), and Deep Space missions. The Johnson Space Center is developing the Environmental Control and Life Support System (ECLSS) for the MMSEV. The MMSEV s intended use is to support longer sortie lengths with multiple Extra Vehicular Activities (EVAs) on a higher magnitude than any previous vehicle. This paper presents an analysis of a high pressure oxygen cascade storage and delivery system that will accommodate the crew during long duration Intra Vehicular Activity (IVA) and capable of multiple high pressure oxygen fills to the Portable Life Support System (PLSS) worn by the crew during EVAs. A cascade is a high pressure gas cylinder system used for the refilling of smaller compressed gas cylinders. Each of the large cylinders are filled by a compressor, but the cascade system allows small cylinders to be filled without the need of a compressor. In addition, the cascade system is useful as a "reservoir" to accommodate low pressure needs. A regression model was developed to provide the mechanism to size the cascade systems subject to constraints such as number of crew, extravehicular activity duration and frequency, and ullage gas requirements under contingency scenarios. The sizing routine employed a numerical integration scheme to determine gas compressibility changes during depressurization and compressibility effects were captured using the Soave-Redlich-Kwong (SRK) equation of state. A multi-dimensional nonlinear optimization routine was used to find the minimum cascade tank system mass that meets the mission requirements. The sizing algorithms developed in this analysis provide a powerful framework to assess cascade filling, compressor, and hybrid systems to design long duration vehicle ECLSS architecture. 1

Yagoda, Evan↗
DOE OSTI2026

Stoichiometrically-informed symbolic regression for extracting chemical reaction mechanisms from data

A data-driven computational method is introduced to extract chemical reaction mechanisms from time series chemical concentration data. It is realized through the use of dynamic symbolic regression in which a sparse analytical form for a dynamical system is discoverable from the underlying data. We specifically develop the stoichiometrically-informed symbolic regression (SISR) method to address a standing challenge in complex chemical reaction networks: given a time-series dataset of concentrations of several components, what is the mechanism and the associated rate constants? SISR finds the optimal mechanism, kinetic equations and rate constants by combining differential optimization with a genetic optimization approach that searches a symbolic space of possible reaction mechanisms. Use of SISR in several paradigmatic examples spanning linear and nonlinear reaction schemes results in excellent agreement between true and predicted mechanisms, including when the method is applied to noisy data. The advantages of a stoichiometrically-informed approach such as SISR to address reaction discovery is illustrated through comparison with the use of generic state-of-the-art data-driven approaches.

36 MATERIALS SCIENCE↗
DOE OSTI2015

Automatic Identification of Closed-Loop Wind Turbine Dynamics via Genetic Programming

Wind turbines are nonlinear systems that operate in turbulent environments. As such, their behavior is difficult to characterize accurately across a wide range of operating conditions by physically meaningful models. Customarily, data-based models of wind turbines are defined in 'black box' format, lacking in both conciseness and physical intelligibility. To address this deficiency, we identify models of a modern horizontal-axis wind turbine in symbolic form using a recently developed symbolic regression method. The method used relies on evolutionary multi-objective optimization to produce succinct dynamic models from operational data without 'a priori' knowledge of the system. We compare the produced models with models derived by other methods for their estimation capacity and evaluate the tradeoff between model intelligibility and accuracy. Several succinct models are found that predict wind turbine behavior as well as or better than more complex alternatives derived by other methods.

17 WIND ENERGY↗
DOE OSTI2020

Multi-objective optimization of peel and shear strengths in ultrasonic metal welding using machine learning-based response surface methodology

Ultrasonic metal welding (UMW) is a solid-state joining technique with varied industrial applications. Despite of its numerous advantages, UMW has a relative narrow operating window and is sensitive to variations in process conditions. As such, it is imperative to quantitatively characterize the influence of welding parameters on the resulting joint quality. The quantification model can be subsequently used to optimize the parameters. Conventional response surface methodology (RSM) usually employs linear or polynomial models, which may not be able to capture the intricate, nonlinear input-output relationships in UMW. Furthermore, some UMW applications call for simultaneous optimization of multiple quality indices such as peel strength, shear strength, electrical conductivity, and thermal conductivity. To address these challenges, this paper develops a machine learning (ML)-based RSM to model the input-output relationships in UMW and jointly optimize two quality indices, namely, peel and shear strengths. The performance of various ML methods including spline regression, Gaussian process regression (GPR), support vector regression (SVR), and conventional polynomial regression models with different orders is compared. A case study using experimental data shows that GPR with radial basis function (RBF) kernel and SVR with RBF kernel achieve the best prediction accuracy. The obtained response surface models are then used to optimize a compound joint strength indicator that is defined as the average of normalized shear and peel strengths. In addition, the case study reveals different patterns in the response surfaces of shear and peel strengths, which has not been systematically studied in the literature. While developed for the UMW application, the method can be extended to other manufacturing processes.

42 ENGINEERING↗
DOE OSTI2021

Beyond optimization—supervised learning applications in relativistic laser-plasma experiments

We explore the applications of machine learning techniques in relativistic laser-plasma experiments beyond optimization purposes. We predict the beam charge of electrons produced in a laser wakefield accelerator given the laser wavefront change caused by a deformable mirror. Machine learning enables feature analysis beyond merely searching for an optimal beam charge, showing that specific aberrations in the laser wavefront are favored in generating higher beam charges. Supervised learning models allow characterizing the measured data quality as well as recognizing irreproducible data and potential outliers. Furthermore, we also include virtual measurement errors in the experimental data to examine the model robustness under these conditions. This work demonstrates how machine learning methods can benefit data analysis and physics interpretation in a highly nonlinear problem of relativistic laser-plasma interaction.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
NASA NTRSDate not supplied

Nonlinear Dynamic Control Derivative Analysis for Aircraft with Application to Transonic Truss-Braced Wing

This paper presents the development of a nonlinear dynamic control derivative estimation method. A nonlinear aerodynamic model is developed to account for the effects of large control surface deflections on aircraft aerodynamic forces and moments. A series of unsteady RANS CFD simulations is performed to simulate the control surface oscillations at various reduced frequencies. The time-domain data are transformed into the frequency-domain data by a Fourier series analysis. Transfer functions of the dynamic control derivatives are then estimated by a frequency-domain regression. The method is applied to the Transonic Truss-Braced Wing (TTBW) to estimate the dynamic control derivatives for the elevator, aileron, and rudder.

Aircraft Control Derivatives↗
DOE OSTI2024

Semi-supervised Learning of Dynamical Systems with Neural Ordinary Differential Equations: A Teacher-Student Model Approach

Modeling dynamical systems is crucial for a wide range of tasks, but it remains challenging due to complex nonlinear dynamics, limited observations, or lack of prior knowledge. Recently, data-driven approaches such as Neural Ordinary Differential Equations (NODE) have shown promising results by leveraging the expressive power of neural networks to model unknown dynamics. However, these approaches often suffer from limited labeled training data, leading to poor generalization and suboptimal predictions. On the other hand, semi-supervised algorithms can utilize abundant unlabeled data and have demonstrated good performance in classification and regression tasks. We propose TS-NODE, the first semi-supervised approach to modeling dynamical systems with NODE. TS-NODE explores cheaply generated synthetic pseudo rollouts to broaden exploration in the state space and to tackle the challenges brought by lack of ground-truth system data under a teacher-student model. TS-NODE employs an unified optimization framework that corrects the teacher model based on the student's feedback while mitigating the potential false system dynamics present in pseudo rollouts. TS-NODE demonstrates significant performance improvements over a baseline Neural ODE model on multiple dynamical system modeling tasks.

Wang, Yu↗
DOE OSTI2022

Estimation of conditional cumulative incidence functions under generalized semiparametric regression models with missing covariates, with application to analysis of biomarker correlates in vaccine trials

Herein, this article presents generalized semiparametric regression models for conditional cumulative incidence functions with competing risks data when covariates are missing by sampling design or happenstance. A doubly robust augmented inverse probability weighted (AIPW) complete-case approach to estimation and inference is investigated. This approach modifies IPW complete-case estimating equations by exploiting the key features in the relationship between the missing covariates and the phase-one data to improve efficiency. An iterative numerical procedure is derived to solve the nonlinear estimating equations. The asymptotic properties of the proposed estimators are established. A simulation study examining the finite-sample performances of the proposed estimators shows that the AIPW estimators are more efficient than the IPW estimators. The developed method is applied to the RV144 HIV-1 vaccine efficacy trial to investigate vaccine-induced IgG binding antibodies to HIV-1 as correlates of acquisition of HIV-1 infection while taking account of whether the HIV-1 sequences are near or far from the HIV-1 sequences represented in the vaccine construct.

97 MATHEMATICS AND COMPUTING↗