Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Regularization”

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 145 records · Page 8
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

GRIDS-Net: Inverse shape design and identification of scatterers via geometric regularization and physics-embedded deep learning

This study presents a deep learning based methodology for both remote sensing and design of acoustic scatterers. The ability to determine the shape of a scatterer, either in the context of material design or sensing, plays a critical role in many practical engineering problems. This class of inverse problems is extremely challenging due to their high-dimensional, nonlinear, and ill-posed nature. To overcome these technical hurdles, we introduce a geometric regularization approach for deep neural networks (DNN) based on non-uniform rational B-splines (NURBS) and capable of predicting complex 2D scatterer geometries in a parsimonious dimensional representation. Then, this geometric regularization is combined with physics-embedded learning and integrated within a robust convolutional autoencoder (CAE) architecture to accurately predict the shape of 2D scatterers in the context of identification and inverse design problems. Further, an extensive numerical study is presented in order to showcase the remarkable ability of this approach to handle complex scatterer geometries while generating physically-consistent acoustic fields. The study also assesses and contrasts the role played by the (weakly) embedded physics in the convergence of the DNN predictions to a physically consistent inverse design.

42 ENGINEERING↗
DOE OSTI2021

A physics-informed and hierarchically regularized data-driven model for predicting fluid flow through porous media

This paper presents a new deep learning data-driven model for predicting structure dependent pore-fluid velocity fields in rock. The model is based on a Convolutional Auto-Encoder (CAE) artificial neural network capable of learning from image data generated by direct numerical simulations of fluid flow through pore-structures, such as by Lattice Boltzmann or molecular dynamics methods. The main novelty of the model in comparison to previous CAE-based data-driven approaches consists of three parts. The first is a methodology for decomposing the full-domain of the porous media into sub-regions, or “sub-domains”, in order to reduce the overall size of the CAE, batch process the sub-domains in parallel, and enable the CAE to learn local and generalizable nonlinear mappings of pore-fluid velocities. The second consists of embedding the finite difference solutions of the incompressible Navier-Stokes and continuity equations into convolutional layers prior to the CAE in order to provide the CAE with knowledge of fluid dynamics physics (PhyFlow). The third main novelty is that the training of the CAE is regularized with a hierarchical loss function that encourages the learning of fluid flow patterns (in a way similar to ranked modes in principal component analysis), ranking from most to least important. This is shown to increase the stability in learning, reduce over-fitting, and promote interpretability of the CAE neural network layers (HierCAE). The comprehensive new data-driven model, which we call the PhyFlow-HierCAE model, is shown to exhibit improved accuracy and generalizability of flow field predictions over conventional CAE models, attributable to the embedded physical knowledge and the hierarchical regularization, as well as realize orders of magnitude speed-ups in computation times as a surrogate for the direct numerical simulations. Examples of training and forward predictions on unseen pore-structures are provided and evaluated for data from Lattice Boltzmann and molecular dynamics simulations of pore-fluid flow. The model is shown to be a fast and accurate emulator (or “surrogate”) for predicting effective permeability of unseen pore-structures based on learning from relatively small direct numerical simulation datasets.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2023

Long–short-term memory encoder–decoder with regularized hidden dynamics for fault detection in industrial processes

The ability of recurrent neural networks (RNN) to model nonlinear dynamics of high dimensional process data has enabled data-driven RNN-based fault detection algorithms. Previous studies have focused on detecting faults by identifying the discrepancies in data distribution between the faulty and normal data, as reflected in prediction errors generated by RNN models. However, in industrial processes, variations in data distribution can also result from changes in normal control setpoints and compensatory control adjustments in response to disturbances, making it hard to differentiate between normal and faulty conditions. This paper proposes a fault detection method utilizing a long short-term memory (LSTM) encoder–decoder structure with regularized hidden dynamics and reversible instance normalization (RevIN) to compactly represent high-dimensional measurements for effective monitoring. During training, the hidden states of the model are regularized to form a low-dimensional latent space representation of the original multivariate time series data. As a result, the prediction errors of the latent states can be used to monitor the abnormal dynamic variations, while the reconstruction errors of the measured variables are used to monitor the abnormal static variations. Furthermore, the proposed indices can reflect operating conditions, even when the distribution of test data changes, which helps distinguish faults from normal adjustments and disturbances that controllers can settle. Here, data from numerical simulation and the Tennessee Eastman process are used to illustrate the effectiveness of the proposed fault detection method.

42 ENGINEERING↗
DOE OSTI2020

A practical way to regularize unfolding of sharply varying spectra with low data statistics

Unfolding is a well-established tool in particle physics. However, a naive application of the standard regularization techniques to unfold the momentum spectrum of protons ejected in the process of negative muon nuclear capture led to a result exhibiting unphysical artifacts. A finite data sample limited the range in which unfolding can be performed, thus introducing a cutoff. A sharply falling “true” distribution led to low data statistics near the cutoff, which exacerbated the regularization bias and produced an unphysical spike in the resulting spectrum. An improved approach has been developed to address these issues and is illustrated using a toy model. The approach uses full Poisson likelihood of data, and produces a continuous, physically plausible, unfolded distribution. Here, the new technique has a broad applicability since sharply falling spectra are common.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

Epistatic Net allows the sparse spectral regularization of deep neural networks for inferring fitness functions

Abstract Despite recent advances in high-throughput combinatorial mutagenesis assays, the number of labeled sequences available to predict molecular functions has remained small for the vastness of the sequence space combined with the ruggedness of many fitness functions. While deep neural networks (DNNs) can capture high-order epistatic interactions among the mutational sites, they tend to overfit to the small number of labeled sequences available for training. Here, we developed Epistatic Net (EN), a method for spectral regularization of DNNs that exploits evidence that epistatic interactions in many fitness functions are sparse. We built a scalable extension of EN, usable for larger sequences, which enables spectral regularization using fast sparse recovery algorithms informed by coding theory. Results on several biological landscapes show that EN consistently improves the prediction accuracy of DNNs and enables them to outperform competing models which assume other priors. EN estimates the higher-order epistatic interactions of DNNs trained on massive sequence spaces-a computational problem that otherwise takes years to solve.

Aghazadeh, Amirali (ORCID:0000000302230873)↗
DOE OSTI2026

Regularization by denoising diffusion models for solving inverse PDE problems with application to full waveform inversion

Partial differential equation (PDE)-governed inverse problems are fundamental across various scientific and engineering applications; yet they face significant challenges due to nonlinearity, ill-posedness, and sensitivity to noise. Here, we introduce a computational framework, regularization by denoising using diffusion models for partial differential equations (RED-DiffEq), by integrating physics-driven inversion and data-driven learning. RED-DiffEq leverages pretrained diffusion models as a regularization mechanism for PDE-governed inverse problems. We apply RED-DiffEq to solve the full waveform inversion problem in geophysics, a challenging seismic imaging technique that seeks to reconstruct high-resolution subsurface velocity models from seismic measurement data. Our method shows enhanced accuracy and robustness compared to benchmark methods. Additionally, it exhibits strong generalization and domain decomposition capacity, enabling the inversion of more complex velocity models with larger domains than those used in training the diffusion model. Our framework can also be directly applied to diverse PDE-governed inverse problems.

Shan, Siming [Yale University, New Haven, CT (Unit↗
DOE OSTI2022

Sequentially regular polymer acceptors featuring flexible spacers for high-performance and mechanically robust all-polymer solar cells

Developing high-performance and mechanically robust polymer solar cells (PSCs) is crucial for realizing wearable power sources. While efficient all-polymer solar cells (all-PSCs) can be fabricated from polymerized small-molecule acceptors (PSMAs) with high optical absorption and electron mobilities, they still show limited mechanical robustness. Here, we achieve highly efficient and mechanically robust all-PSCs by designing a PSMA (PYFS-Reg) containing sequence-regular flexible spacers (FSs). The regular incorporation of the FS units into PSMAs is essential in simultaneously improving the electrical and mechanical properties of blend films. As a result, all-PSCs featuring PYFS-Reg achieve a high power conversion efficiency (PCE = 16.1%) and stretchability (crack onset strain (COS) = 22.4%), outperforming PSMAs without FSs (i.e., PYBDT, PCE = 12.6% and COS = 11.7%) or with randomly distributed FSs (i.e., PYFS-Ran, PCE = 12.2% and COS = 18.1%). Importantly, these all-PSCs are fabricated by an environmentally benign, non-halogenated solvent process. To further demonstrate their feasible applications in wearable devices, we construct intrinsically stretchable (IS) all-PSCs by using PYFS-Reg-based active layers, which exhibit a high PCE (10.6%) and excellent device stretchability (strain at PCE 80% = 36.7%).

14 SOLAR ENERGY↗
DOE OSTI2016

Mutual interaction between high and low stereo-regularity components for crystallization and melting behaviors of polypropylene blend fibers

Crystallization and melting behaviors of blend fibers of two types of polypropylene (PP), i.e. high stereo-regularity/high molecular weight PP (HPP) and low stereo-regularity/low molecular weight PP (LPP), was investigated. Blend fibers consisting of various HPP/LPP compositions were prepared through the melt spinning process. Differential scanning calorimetry (DSC), temperature modulated DSC (TMDSC) and wide-angle X-ray diffraction (WAXD) analysis were applied for clarifying the crystallization and melting behaviors of individual components. In the DSC measurement of blend fibers with high LPP composition, continuous endothermic heat was detected between the melting peaks of LPP at around 40 °C and that of HPP at around 160 °C. Such endothermic heat was more distinct for the blend fibers with higher LPP composition indicating that the melting of LPP in the heating process was hindered because of the presence of HPP crystals. On the other hand, heat of crystallization was detected at around 90 °C in the case of blend fibers with LPP content of 30 to 70 wt%, indicating that the crystallization of HPP component was taking place during the heating of as-spun blend fibers in the DSC measurement. Through the TMDSC analysis, re-organization of the crystalline structure through the simultaneous melting and re-crystallization was detected in the cases of HPP and blend fibers, whereas re-crystallization was not detected during the melting of LPP fibers. In the WAXD analysis during the heating of fibers, amount of a-form crystal was almost constant up to the melting in the case of single component HPP fibers, whereas there was a distinct increase of the intensity of crystalline reflections from around 100 °C, right after the melting of LPP in the case of blend fibers. These results suggested that the crystallization of HPP in the spinning process as well as during the conditioning process after spinning was hindered by the presence of LPP.

36 MATERIALS SCIENCE↗
DOE OSTI2020

A light weight regularization for wave function parameter gradients in quantum Monte Carlo

The parameter derivative of the expectation value of the energy, ∂E/∂p, is a key ingredient in variational Monte Carlo (VMC) wave function optimization methods. In some cases, a naïve estimate of this derivative suffers from an infinite variance, which inhibits the efficiency of optimization methods that rely on a stable estimate of the derivative. In this work, we derive a simple regularization of the naïve estimator, which is trivial to implement in existing VMC codes, has finite variance, and a negligible bias, which can be extrapolated to zero bias with no extra cost. We use this estimator to construct an unbiased, finite variance estimation of ∂E/∂p for a multi-Slater–Jastrow trial wave function on the LiH molecule and in the optimization of a multi-Slater–Jastrow trial wave function on the CuO molecule. This regularized estimator is a simple and efficient estimator of ∂E/∂p for VMC optimization techniques.

36 MATERIALS SCIENCE↗
DOE OSTI2016

Adiabatic regularization of power spectra in nonminimally coupled chaotic inflation

We investigate the effect of adiabatic regularization on both the tensor- and scalar-perturbation power spectra in nonminimally coupled chaotic inflation. Similar to that of the minimally coupled general single-field inflation, we find that the subtraction term is suppressed by an exponentially decaying factor involving the number of e -folds. By following the subtraction term long enough beyond horizon crossing, the regularized power spectrum tends to the ''bare'' power spectrum. This study justifies the use of the unregularized (''bare'') power spectrum in standard calculations.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

Acoustic- and elastic-waveform inversion with total generalized p-variation regularization

Geophysical models usually contain both sharp interfaces and smooth variations, and it is difficult to accurately account for both of these two types of medium parameter variations using conventional full-waveform inversion methods. We develop a novel full-waveform inversion method for acoustic and elastic waves using a total generalized p-variation regularization scheme to address these challenging problems. We decompose the full-waveform inversion into two subproblems and solve these two minimization subproblems using an alternating-direction minimization strategy. One significant advantage of the total generalized p-variation regularization scheme is that it can simultaneously reconstruct sharp interfaces and smooth background variations of geophysical parameters. Such capability can also effectively suppress the noises in source-encoded inversion and sparse-data inversion, or inversion of noisy data. We demonstrate the advantages of our full-waveform inversion method using a checkerboard model, a modified elastic SEG/EAGE overthrust model, and a land field seismic data set. Our conclusions of synthetic and field seismic data demonstrate that our method reconstructs both smooth background variations and sharp interfaces of subsurface geophysical properties accurately, and provides a useful tool for accurate and reliable inversion of field seismic data.

58 GEOSCIENCES↗
DOE OSTI2025

Monomer-dimer tensor-network basis for qubit-regularized lattice gauge theories

Traditional SU⁡(𝑁) lattice gauge theories (LGTs) can be formulated using an orthonormal basis constructed from the irreducible representations (irreps) 𝑉 𝜆 of the SU⁡(𝑁) gauge symmetry. On a lattice, the elements of this basis are tensor networks comprising dimer tensors on the links labeled by a set of irreps {𝜆 ℓ } and monomer tensors on sites labeled by {𝜆 𝑠 }. These tensors naturally define a local site Hilbert space, ℋ$^𝑔_𝑠$, on which gauge transformations act. Gauss’s law introduces an additional index 𝛼 𝑠 =1,2,…,𝒟⁡(ℋ$^𝑔_𝑠$) that labels an orthonormal basis of the gauge-invariant subspace of ℋ$^𝑔_𝑠$. This monomer-dimer tensor-network (MDTN) basis, |{𝜆 𝑠 },{𝜆 ℓ },{𝛼 𝑠 }⟩, of the physical Hilbert space enables the construction of new qubit-regularized SU⁡(𝑁) gauge theories that are free of sign problems while preserving key features of traditional LGTs. Here, we investigate finite-temperature confinement-deconfinement transitions in a simple qubit-regularized SU(2) and SU(3) gauge theory in 𝑑 =2 and 𝑑 =3 spatial dimensions, formulated using the MDTN basis, and show that they reproduce the universal results of traditional LGTs at these transitions. Additionally, in 𝑑 =1, we demonstrate using a plaquette chain that the string tension at zero temperature can be continuously tuned to zero by adjusting a model parameter that plays the role of the gauge coupling in traditional LGTs.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Asymptotic Freedom at the Berezinskii-Kosterlitz-Thouless Transition without Fine-Tuning Using a Qubit Regularization

We propose a two-dimensional hard-core loop-gas model as a way to regularize the asymptotically free massive continuum quantum field theory that emerges at the Berezinskii-Kosterlitz-Thouless transition. Without fine-tuning, our model can reproduce the universal step-scaling function of the classical lattice X Y model in the massive phase as we approach the phase transition. This is achieved by lowering the fugacity of Fock-vacuum sites in the loop-gas configuration space to zero in the thermodynamic limit. Some of the universal quantities at the Berezinskii-Kosterlitz-Thouless transition show smaller finite size effects in our model as compared to the traditional X Y model. Our model is a prime example of qubit regularization of an asymptotically free massive quantum field theory in Euclidean space-time and helps understand how asymptotic freedom can arise as a relevant perturbation at a decoupled fixed point without fine-tuning. Published by the American Physical Society 2024

Physics↗
DOE OSTI2026

Regularization prescription for the mixing between nonlocal gluon and quark operators

It is well known that in the study of mixing between nonlocal gluon and quark bilinear operators there exists an ambiguity when relating coordinate space and momentum space results. In this work, we show that this ambiguity is due to the lack of a proper regularization prescription of the singularity that arises when the separation between the gluon/quark fields approaches zero. We then demonstrate that dimensional regularization resolves this issue and yields consistent results in both coordinate and momentum space. This prescription is also compatible with lattice extractions of parton distributions from nonlocal operators.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Source function from two-particle correlation function through entropy-regularized Richardson-Lucy deblurring

Source functions are obtained from p – p and d – α correlation functions by applying the Richardson-Lucy (RL) deblurring to the Koonin-Pratt (KP) equation. To prevent fitting of noise in the correlation function, total-variation (TV) regularization is employed that has been effective in ordinary image restoration. TV alone cannot ensure normalization of the source functions. To ensure the latter, we propose a maximum-entropy regularized RL algorithm (MEM-RL). We outline the MEM-RL formalism and optimization strategy for the KP equation, demonstrating its effectiveness on both simulated and experimental data, including the p – p and d – α correlation functions.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2024

RADEMACHER COMPLEXITY REGULARIZATION FOR CORRELATION-BASED MULTIVIEW REPRESENTATION LEARNING

Deep correlation-based multiview representation learning techniques have become increasingly popular methods for extracting highly correlated representations from multiview data. However, their ability to find highly complex mappings between the views can also lead to overfitting and overly correlated representations. In this work, we propose a regularizer for this specific problem, based on the Rademacher complexity of the DNNs, tailored for multiview correlation maximization. We demonstrate that the proposed regularization leads to less noisy representations in synthetic data and improved performance of downstream tasks in real-world multiview datasets.

Kuschel, Maurice↗
DOE OSTI2021

srMO-BO-3GP: A sequential regularized multi-objective Bayesian optimization for constrained design applications using an uncertain Pareto classifier

Bayesian optimization (BO) is an efficient and flexible global optimization framework that is applicable to a very wide range of engineering applications. To leverage the capability of the classical BO, many extensions, including multi-objective, multi-fidelity, parallelization, and latent-variable modeling, have been proposed to address the limitations of the classical BO framework. In this work, we propose a novel multi-objective BO formalism, called srMO-BO-3GP, to solve multi-objective optimization problems in a sequential setting. Three different Gaussian processes (GPs) are stacked together, where each of the GPs is assigned with a different task. The first GP is used to approximate a single-objective computed from the multi-objective definition, the second GP is used to learn the unknown constraints, and the third one is used to learn the uncertain Pareto frontier. At each iteration, a multi-objective augmented Tchebycheff function is adopted to convert multi-objective to single-objective, where the regularization with a regularized ridge term is also introduced to smooth the single-objective function. Finally, we couple the third GP along with the classical BO framework to explore the convergence and diversity of the Pareto frontier by the acquisition function for exploitation and exploration. The proposed framework is demonstrated using several numerical benchmark functions, as well as a thermomechanical finite element model for flip-chip package design optimization.

42 ENGINEERING↗
DOE OSTI2016

Temporal regularization of ultrasound-based liver motion estimation for image-guided radiation therapy

Purpose: Ultrasound-based motion estimation is an expanding subfield of image-guided radiation therapy. Although ultrasound can detect tissue motion that is a fraction of a millimeter, its accuracy is variable. For controlling linear accelerator tracking and gating, ultrasound motion estimates must remain highly accurate throughout the imaging sequence. This study presents a temporal regularization method for correlation-based template matching which aims to improve the accuracy of motion estimates. Methods: Liver ultrasound sequences (15–23 Hz imaging rate, 2.5–5.5 min length) from ten healthy volunteers under free breathing were used. Anatomical features (blood vessels) in each sequence were manually annotated for comparison with normalized cross-correlation based template matching. Five sequences from a Siemens Acuson™ scanner were used for algorithm development (training set). Results from incremental tracking (IT) were compared with a temporal regularization method, which included a highly specific similarity metric and state observer, known as the α–β filter/similarity threshold (ABST). A further five sequences from an Elekta Clarity™ system were used for validation, without alteration of the tracking algorithm (validation set). Results: Overall, the ABST method produced marked improvements in vessel tracking accuracy. For the training set, the mean and 95th percentile (95%) errors (defined as the difference from manual annotations) were 1.6 and 1.4 mm, respectively (compared to 6.2 and 9.1 mm, respectively, for IT). For each sequence, the use of the state observer leads to improvement in the 95% error. For the validation set, the mean and 95% errors for the ABST method were 0.8 and 1.5 mm, respectively. Conclusions: Ultrasound-based motion estimation has potential to monitor liver translation over long time periods with high accuracy. Nonrigid motion (strain) and the quality of the ultrasound data are likely to have an impact on tracking performance. A future study will investigate spatial uniformity of motion and its effect on the motion estimation errors.

61 RADIATION PROTECTION AND DOSIMETRY↗