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 361 records · Page 20

Non-intrusive nonlinear model reduction via machine learning approximations to low-dimensional operators

Abstract Although projection-based reduced-order models (ROMs) for parameterized nonlinear dynamical systems have demonstrated exciting results across a range of applications, their broad adoption has been limited by their intrusivity: implementing such a reduced-order model typically requires significant modifications to the underlying simulation code. To address this, we propose a method that enables traditionally intrusive reduced-order models to be accurately approximated in a non-intrusive manner. Specifically, the approach approximates the low-dimensional operators associated with projection-based reduced-order models (ROMs) using modern machine-learning regression techniques. The only requirement of the simulation code is the ability to export the velocity given the state and parameters; this functionality is used to train the approximated low-dimensional operators. In addition to enabling nonintrusivity, we demonstrate that the approach also leads to very low computational complexity, achieving up to $$10^3{\times }$$ 10 3 × in run time. We demonstrate the effectiveness of the proposed technique on two types of PDEs. The domain of applications include both parabolic and hyperbolic PDEs, regardless of the dimension of full-order models (FOMs).

42 ENGINEERING↗

A Parametric, Data-Driven, Non-Intrusive Reduced-Order Model Framework for Crystal Plasticity Simulations of Voids

The influence of the internal structure at micrometer length scales on the deformation of polycrystalline materials can be effectively captured using crystal plasticity finite element methods (CPFEM). However, the complexity and nonlinearity of the deformation equations CPFEM solves demand significant computational power and resources to achieve accurate predictions, limiting its broader application. To address this challenge, we have identified a reduced-order representation of the complex data in order to establish a computationally efficient reduced-order models (ROM) and drastically reduce the computational expense of CPFEM. Specifically, in this work, we developed a parametric, data-driven, and non-intrusive ROM framework for CPFEM using proper orthogonal decomposition (POD) and sparse variational Gaussian process (SVGP) regression for single-crystal microstructures under tensile loading conditions. The developed protocol enables one to compress field into a latent/low-dimensional space described by principal component analysis (PCA) via the singular value decomposition (SVD) algorithm. As a result, the high-dimensional data are reduced to a significantly smaller amount of dimensions with POD bases and POD coefficients. Furthermore, we deployed an ensemble of SVGPs—extended from the classical Gaussian process (GP) regression for scalability and handling big data—in a massively parallel manner to train and predict latent POD coefficients using known POD bases from a set of previously obtained simulations results. Lastly, using the predicted POD coefficients, we reconstructed the full-field results and showed reasonable agreement compared with the true values obtained from running CPFEM. The developed framework is validated with a set of CPFEM simulations of a single embedded void in single-crystal aluminum alloy. While the framework is broadly applicable, this work specifically focuses on single-crystal microstructures, a single load case (e.g., tensile), and a specific void geometry (spherical).

Anisotropy↗

Machine learning models for estimating contamination across different curbside collection strategies

Contaminated recyclables, which are frequently discarded as waste, pose a significant challenge to the implementation of a circular economy. These contaminated recyclables impede the circulation of resources, resulting in higher processing costs at material recovery facilities (MRFs). Over the past few decades, machine learning (ML) models such as linear regression (LR), support vector machine (SVM), and random forest (RF) have evolved to provide new methods for predicting inbound contamination rates in addition to traditional statistical models. In this study, we applied ML models to predict inbound contamination rates using demographic features from 15 counties in the U.S. with different curbside collection strategies. In general, we found that ML models outperformed linear mixed models. Specifically, SVM models had the highest performance (R 2 = 0.75; mean absolute error (MAE) = 0.06), which may be due to their ability to model nonlinear relationships between features and inbound contamination rates. Further, the key predictor was population, with poverty rate being positively correlated and median age negatively correlated with inbound contamination rates. To improve the management of contamination and enhance the implementation of a circular economy, better models are needed to understand and estimate inbound contamination rates as well as identify critical factors in the present and future.

54 ENVIRONMENTAL SCIENCES↗

Chemiluminescence-based multivariate sensing of local equivalence ratios in premixed atmospheric methane-air flames

Chemiluminescence emissions from OH*, CH*, C2, and CO2 formed within the reaction zone of premixed flames depend upon the fuel-air equivalence ratio in the burning mixture. In the present paper, a new partial least square regression (PLS-R) based multivariate sensing methodology is investigated and compared with an OH*/CH* intensity ratio-based calibration model for sensing equivalence ratio in atmospheric methane-air premixed flames. Five replications of spectral data at nine different equivalence ratios ranging from 0.73 to 1.48 were used in the calibration of both models. During model development, the PLS-R model was initially validated with the calibration data set using the leave-one-out cross validation technique. Since the PLS-R model used the entire raw spectral intensities, it did not need the nonlinear background subtraction of CO2 emission that is required for typical OH*/CH* intensity ratio calibrations. An unbiased spectral data set (not used in the PLS-R model development), for 28 different equivalence ratio conditions ranging from 0.71 to 1.67, was used to predict equivalence ratios using the PLS-R and the intensity ratio calibration models. It was found that the equivalence ratios predicted with the PLS-R based multivariate calibration model matched the experimentally measured equivalence ratios within 7%; whereas, the OH*/CH* intensity ratio calibration grossly underpredicted equivalence ratios in comparison to measured equivalence ratios, especially under rich conditions ( > 1.2). The practical implications of the chemiluminescence-based multivariate equivalence ratio sensing methodology are also discussed.

Cross validation↗

Latent space dynamics identification for interface tracking with application to shock-induced pore collapse

Capturing sharp, evolving interfaces remains a central challenge in reduced-order modeling, especially when data is limited and the system exhibits localized nonlinearities or discontinuities. Here, we propose LaSDI-IT (Latent Space Dynamics Identification for Interface Tracking), a data-driven framework that combines low-dimensional latent dynamics learning with explicit interface-aware encoding to enable accurate and efficient modeling of physical systems involving moving material boundaries. At the core of LaSDI-IT is a revised autoencoder architecture that jointly reconstructs the physical field and an indicator function representing material regions or phases, allowing the model to track complex interface evolution without requiring detailed physical models or mesh adaptation. The latent dynamics are learned through linear regression in the encoded space and generalized across parameter regimes using Gaussian process interpolation with greedy sampling. We demonstrate LaSDI-IT on the problem of shock-induced pore collapse in high explosives, a process characterized by sharp temperature gradients and dynamically deforming pore geometries. The method achieves relative prediction errors below 9% across the parameter space, accurately recovers key quantities of interest such as pore area and hot spot formation, and matches the performance of dense training with only half the data. This latent dynamics prediction was 10 6 times faster than the conventional high-fidelity simulation, proving its utility for multi-query applications. These results highlight LaSDI-IT as a general, data-efficient framework for modeling discontinuity-rich systems in computational physics, with potential applications in multiphase flows, fracture mechanics, and phase change problems.

Gaussian process↗

Survival of Endangered Snake River Sockeye Salmon Smolts from Three Idaho Lakes: Relationships with Parr Size at Release, Parr Growth Rate, Smolt Size, Discharge, and Travel Time

Abstract In 1991, Snake River sockeye salmon Oncorhynchus nerka were listed as endangered. The Sawtooth Valley Project was initiated to conserve and rebuild sockeye salmon populations that historically spawned and reared in five Sawtooth Valley lakes designated as critical habitat in central Idaho. We evaluated smolt survival of sockeye salmon that were stocked as parr into Redfish, Pettit, and Alturas lakes. Smolt travel time, residuals of Salmon River discharge and smolt travel time, and specific growth rate of parr explained 58% of the variation in smolt survival in a multiple-regression model. Smolt survival was inversely related to smolt travel time, and travel time was negatively correlated with mean May discharge in the Salmon River at Salmon, Idaho. We were particularly interested in the relationships between smolt survival and parr size at release, smolt size at migration, and parr growth rate. Smolt survival from nursery lakes to Lower Granite Dam was negatively correlated with mean parr size at release and mean smolt size at emigration. Smaller parr and smolts survived better. Smolt survival to the dam was correlated with parr growth rates in the three lakes combined; the relationship was nonlinear. Smolt survival increased as specific growth rate increased up to 0.06% per day, but further increases in growth rate were associated with reductions in survival. Absolute growth rates of parr were negatively related to parr weight at release. Smaller individuals grew faster than larger individuals, apparently as a result of size-dependent metabolic demand and prey availability. The relationships between smolt survival and parr size, smolt size, and parr growth rate differed among lakes. These data suggest that successful migrants must at least maintain weight during the winter preceding migration and that the stocking of smaller parr with lower metabolic demand may be preferable to stocking larger parr when forage is limited. Received November 18, 2010; accepted June 1, 2011

Griswold, Robert G.↗

Application of regression, variance, and density based global sensitivity methods to integrated VERA-CS and BISON simulations [Application of regression, variance, and density based global sensitivity methods to integrated VERA-CS and BISON simulations]

Uncertainty quantification (UQ) and sensitivity analyses (SA) are performed for coupled simulations between VERA-CS, a coupled pin resolved neutron transport and subchannel thermal hydraulics code, and the fuel performance code BISON. The interface between VERA-CS and BISON is performed in a multiphysics environment known as the LOCA Toolkit for U.S. light water reactors (LOTUS) currently under development at Idaho National Laboratory (INL). A focus is placed on using a variety of SA measures, including two regression based (Pearson and Spearman), one variance based (Sobol indices), and three moment independent measures (Delta moment independent measures with L 1 , L 2 , and L ∞ norms). The problem under inspection is a single assembly depletion case for three fuel cycles. The figures of merit are the minimum departure from nucleate boiling ratio (MDNBR), maximum fuel centerline temperature (MFCT), and gap conductance at peak power (GCPP). SA results show MDNBR to be linear with consistent rankings throughout the fuel cycles. MFCT is linear, but with a change in rankings at the switch from open gap to closed gap models. GCPP is nonlinear at intermediate states that coincide with the onset of contact between fuel and cladding. Furthermore, these nonlinear states allow for the showcasing of higher order SA measures over first order methods.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Cardiac interbeat interval dynamics from childhood to senescence : comparison of conventional and new measures based on fractals and chaos theory

BACKGROUND: New methods of R-R interval variability based on fractal scaling and nonlinear dynamics ("chaos theory") may give new insights into heart rate dynamics. The aims of this study were to (1) systematically characterize and quantify the effects of aging from early childhood to advanced age on 24-hour heart rate dynamics in healthy subjects; (2) compare age-related changes in conventional time- and frequency-domain measures with changes in newly derived measures based on fractal scaling and complexity (chaos) theory; and (3) further test the hypothesis that there is loss of complexity and altered fractal scaling of heart rate dynamics with advanced age. METHODS AND RESULTS: The relationship between age and cardiac interbeat (R-R) interval dynamics from childhood to senescence was studied in 114 healthy subjects (age range, 1 to 82 years) by measurement of the slope, beta, of the power-law regression line (log power-log frequency) of R-R interval variability (10(-4) to 10(-2) Hz), approximate entropy (ApEn), short-term (alpha(1)) and intermediate-term (alpha(2)) fractal scaling exponents obtained by detrended fluctuation analysis, and traditional time- and frequency-domain measures from 24-hour ECG recordings. Compared with young adults (<40 years old, n=29), children (<15 years old, n=27) showed similar complexity (ApEn) and fractal correlation properties (alpha(1), alpha(2), beta) of R-R interval dynamics despite lower spectral and time-domain measures. Progressive loss of complexity (decreased ApEn, r=-0.69, P<0.001) and alterations of long-term fractal-like heart rate behavior (increased alpha(2), r=0.63, decreased beta, r=-0.60, P<0.001 for both) were observed thereafter from middle age (40 to 60 years, n=29) to old age (>60 years, n=29). CONCLUSIONS: Cardiac interbeat interval dynamics change markedly from childhood to old age in healthy subjects. Children show complexity and fractal correlation properties of R-R interval time series comparable to those of young adults, despite lower overall heart rate variability. Healthy aging is associated with R-R interval dynamics showing higher regularity and altered fractal scaling consistent with a loss of complex variability.

NASA Discipline Cardiopulmonary↗

Comparison of U.S. and Russian Slow Crack Growth Data and Models

Both the US and Russian space programs use similar predictive models for design of fused silica windows on the International space station. The Russian model can be derived from the power expression for slow crack growth (SCG) or “static fatigue.” The US uses both power and exponential models. Despite the similarity of models and data fitting approach (linear regression and right censoring), different SCG parameters have been derived by US and Russian parties for the same material (Russian fused silica) tested in a similar manner. The difference appears to be related to the use of short-term strength data along with the longer-term static fatigue data, with the power law parameter n being very sensitive to the conversion of strength data into equivalent static data. This hybrid approach is feasible if strength data is measured with a constant stress rate and is appropriately converted to a static equivalent. More research into the approach is needed. However, because of the nonlinear behavior of fused silica in log(v) – log(K(I) ) space and the sensitivity of parameter estimation to fit range, the exponential model is a better choice regardless of test method. Functions are given to convert parameters from the Russian model to those in the US model. However, accurate conversion is hampered by the lack of inert strength data. When the same test technique is used, US and Russian materials exhibit very similar parameters.

Silica, strength, impact, crack growth, windows, I↗

Function Preserving Projection

With function preserving projection (FPP), we aim to address high-dimensional data exploration challenges common in many scientific applications. Instead of preserving the global or local geometric structure of a dataset like traditional dimensionality reduction methods (e.g., principal component analysis, t-stochastic neighborhood embedding), the proposed approach finds a linear projection that reveal interpretable (yet potentially nonlinear) pattern of the properties of interests defined on a highdimensional domain (e.g., simulation yield defined on the input parameter space). The key insight driving FPP is the fact that while non-linear embeddings are difficult to interpret, humans are highly skilled in understanding complex visual patterns. FPP lets users select the complexity of a pattern they consider interpretable, allows to find a joint projection that explains multiple functions, and provides a confidence statistic (akin to a p-value) indicating how salient the resulting pattern is. We demonstrate the efficacy of the proposed method on several synthetic and scientific datasets for both continuous (regression) and discrete (classification) problems of interests and illustrate the scalability of FPP by finding an optimal projection for close to 40 gigabytes in around 10 minutes.

Liu, SHUSEN↗

Deep transfer operator learning for partial differential equations under conditional shift

Transfer learning enables the transfer of knowledge gained while learning to perform one task (source) to a related but different task (target), hence addressing the expense of data acquisition and labelling, potential computational power limitations and dataset distribution mismatches. Here, we propose a new transfer learning framework for task-specific learning (functional regression in partial differential equations) under conditional shift based on the deep operator network (DeepONet). Task-specific operator learning is accomplished by fine-tuning task-specific layers of the target DeepONet using a hybrid loss function that allows for the matching of individual target samples while also preserving the global properties of the conditional distribution of the target data. Inspired by conditional embedding operator theory, we minimize the statistical distance between labelled target data and the surrogate prediction on unlabelled target data by embedding conditional distributions onto a reproducing kernel Hilbert space. We demonstrate the advantages of our approach for various transfer learning scenarios involving nonlinear partial differential equations under diverse conditions due to shifts in the geometric domain and model dynamics. Our transfer learning framework enables fast and efficient learning of heterogeneous tasks despite considerable differences between the source and target domains.

42 ENGINEERING↗

A hybrid neurocomputing/numerical strategy for nonlinear structural analysis

A hybrid neurocomputing/numerical strategy is presented for geometrically nonlinear analysis of structures. The strategy combines model-free data processing capabilities of computational neural networks with a Pade approximants-based perturbation technique to predict partial information about the nonlinear response of structures. In the hybrid strategy, multilayer feedforward neural networks are used to extend the validity of solutions by using training samples produced by Pade approximations to the Taylor series expansion of the response function. The range of validity of the training samples is taken to be the radius of convergence of Pade approximants and is estimated by setting a tolerance on the diverging approximants. The norm of residual vector of unbalanced forces in a given element is used as a measure to assess the quality of network predictions. To further increase the accuracy and the range of network predictions, additional training data are generated by either applying linear regression to weight matrices or expanding the training data by using predicted coefficients in a Taylor series. The effectiveness of the hybrid strategy is assessed by performing large-deflection analysis of a doubly-curved composite panel with a circular cutout, and postbuckling analyses of stiffened composite panels subjected to an in-plane edge shear load. In all the problems considered, the hybrid strategy is used to predict selective information about the structural response, namely the total strain energy and the maximum displacement components only.

Szewczyk, Z. Peter↗

Identifying Differential Equations in Fourier Domain (FourierIdent)

We investigate identifying differential equations in the frequency domain. Fourier analysis is an important tool in theoretical analysis and numerical solvers of differential equations, yet there is limited work in exploring this connection in the identification of differential equations. This paper aims to identify the underlying differential equation in the frequency domain, from a given single realization of the differential equation perturbed by noise. Such setting imposes difficulties which are different from other identification methods where computation is carried out in the physical domain. We propose several ways to mitigate the challenges arising from noise in data and large differences in the magnitudes of frequency responses. The main takeaways are that identifying differential equations solely in the frequency domain is challenging, the method we propose is based on a form of domain partitions in the frequency domain, and this method shows benefits for complex data even with high level of noise. We introduce a Fourier feature denoising, and define the meaningful data region and the core regions of features to reduce the effect of noise in the frequency domain and to enhance the accuracy in coefficient identification. The proposed method is tested on various differential equations with linear, nonlinear, and high-order derivative feature terms, and shows advantages on complex data with many frequency modes, even under high level of noise.

97 MATHEMATICS AND COMPUTING↗

Reduced models for ETG transport in the tokamak pedestal

This paper reports on the development of reduced models for electron temperature gradient (ETG) driven transport in the pedestal. Model development is enabled by a set of 61 nonlinear gyrokinetic simulations with input parameters taken from pedestals in a broad range of experimental scenarios. The simulation data have been consolidated in a new database for gyrokinetic simulation data, the multiscale gyrokinetic database (MGKDB), facilitating the analysis. The modeling approach may be considered a generalization of the standard quasilinear mixing length procedure. The parameter η, the ratio of the density to temperature gradient scale length, emerges as the key parameter for formulating an effective saturation rule. With a single order-unity fitting coefficient, the model achieves an error of 15%. A similar model for ETG particle flux is also described. We also present simple algebraic expressions for the transport informed by an algorithm for symbolic regression.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Space Shuttle Main Engine performance analysis

For a number of years, NASA has relied primarily upon periodically updated versions of Rocketdyne's power balance model (PBM) to provide space shuttle main engine (SSME) steady-state performance prediction. A recent computational study indicated that PBM predictions do not satisfy fundamental energy conservation principles. More recently, SSME test results provided by the Technology Test Bed (TTB) program have indicated significant discrepancies between PBM flow and temperature predictions and TTB observations. Results of these investigations have diminished confidence in the predictions provided by PBM, and motivated the development of new computational tools for supporting SSME performance analysis. A multivariate least squares regression algorithm was developed and implemented during this effort in order to efficiently characterize TTB data. This procedure, called the 'gains model,' was used to approximate the variation of SSME performance parameters such as flow rate, pressure, temperature, speed, and assorted hardware characteristics in terms of six assumed independent influences. These six influences were engine power level, mixture ratio, fuel inlet pressure and temperature, and oxidizer inlet pressure and temperature. A BFGS optimization algorithm provided the base procedure for determining regression coefficients for both linear and full quadratic approximations of parameter variation. Statistical information relative to data deviation from regression derived relations was also computed. A new strategy for integrating test data with theoretical performance prediction was also investigated. The current integration procedure employed by PBM treats test data as pristine and adjusts hardware characteristics in a heuristic manner to achieve engine balance. Within PBM, this integration procedure is called 'data reduction.' By contrast, the new data integration procedure, termed 'reconciliation,' uses mathematical optimization techniques, and requires both measurement and balance uncertainty estimates. The reconciler attempts to select operational parameters that minimize the difference between theoretical prediction and observation. Selected values are further constrained to fall within measurement uncertainty limits and to satisfy fundamental physical relations (mass conservation, energy conservation, pressure drop relations, etc.) within uncertainty estimates for all SSME subsystems. The parameter selection problem described above is a traditional nonlinear programming problem. The reconciler employs a mixed penalty method to determine optimum values of SSME operating parameters associated with this problem formulation.

Santi, L. Michael↗

Latent-space time evolution of non-intrusive reduced-order models using Gaussian process emulation

Non-intrusive reduced-order models (ROMs) have recently generated considerable interest for constructing computationally efficient counterparts of nonlinear dynamical systems emerging from various domain sciences. They provide a low-dimensional emulation framework for systems that may be intrinsically high-dimensional. This is accomplished by utilizing a construction algorithm that is purely data-driven. It is no surprise, therefore, that the algorithmic advances of machine learning have led to non-intrusive ROMs with greater accuracy and computational gains. However, in bypassing the utilization of an equation-based evolution, it is often seen that the interpretability of the ROM framework suffers. This becomes more problematic when black-box deep learning methods are used which are notorious for lacking robustness outside the physical regime of the observed data. In this article, we propose the use of a novel latent-space interpolation algorithm based on Gaussian process regression. Notably, this reduced-order evolution of the system is parameterized by control parameters to allow for interpolation in space. The use of this procedure also allows for a continuous interpretation of time which allows for temporal interpolation. The latter aspect provides information, with quantified uncertainty, about full-state evolution at a finer resolution than that utilized for training the ROMs. This research assesses the viability of this algorithm for an advection-dominated system given by the inviscid shallow water equations.

97 MATHEMATICS AND COMPUTING↗

Correlation-Aided Robust Decentralized Dynamic State Estimation of Power Systems with Unknown Control Inputs

This paper proposes a correlation-aided robust adaptive unscented Kalman filter for power system decentralized dynamic state estimation with unknown inputs, termed as robust AUKF-UI. The temporal and spatial correlations among the unknown inputs are used to derive a vector auto-regressive (VAR) model in an adaptive manner. This VAR model is further integrated together with state transition and measurement models for joint state and unknown inputs estimation. This allows taking into account the implicit cross-correlations between the states and the unknown inputs. As a result, the rank requirement for unknown input vector estimation is relaxed and the local generator frequency measurement is not required. The temporal correlations of time series innovation vectors, predicted state and input vectors are also leveraged by the robust AUKFUI to detect, identify and process bad data. Without these correlations, it is very challenging to address bad data with unknown inputs. Simulation results carried out on the IEEE 39-bus system demonstrate that the proposed robust AUKF-UI achieves much better results than other methods in the presence of low measurement redundancy, strong nonlinearity, and bad data.

Zhao, Junbo↗