Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “optimal accuracy”

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 973 records · Page 54

SU-G-JeP3-01: A Method to Quantify Lung SBRT Target Localization Accuracy Based On Digitally Reconstructed Fluoroscopy

Purpose: To develop a methodology based on digitally-reconstructed-fluoroscopy (DRF) to quantitatively assess target localization accuracy of lung SBRT, and to evaluate using both a dynamic digital phantom and a patient dataset. Methods: For each treatment field, a 10-phase DRF is generated based on the planning 4DCT. Each frame is pre-processed with a morphological top-hat filter, and corresponding beam apertures are projected to each detector plane. A template-matching algorithm based on cross-correlation is used to detect the tumor location in each frame. Tumor motion relative beam aperture is extracted in the superior-inferior direction based on each frame’s impulse response to the template, and the mean tumor position (MTP) is calculated as the average tumor displacement. The DRF template coordinates are then transferred to the corresponding MV-cine dataset, which is retrospectively filtered as above. The treatment MTP is calculated within each field’s projection space, relative to the DRF-defined template. The field’s localization error is defined as the difference between the DRF-derived-MTP (planning) and the MV-cine-derived-MTP (delivery). A dynamic digital phantom was used to assess the algorithm’s ability to detect intra-fractional changes in patient alignment, by simulating different spatial variations in the MV-cine and calculating the corresponding change in MTP. Inter-and-intra-fractional variation, IGRT accuracy, and filtering effects were investigated on a patient dataset. Results: Phantom results demonstrated a high accuracy in detecting both translational and rotational variation. The lowest localization error of the patient dataset was achieved at each fraction’s first field (mean=0.38mm), with Fx3 demonstrating a particularly strong correlation between intra-fractional motion-caused localization error and treatment progress. Filtering significantly improved tracking visibility in both the DRF and MV-cine images. Conclusion: We have developed and evaluated a methodology to quantify lung SBRT target localization accuracy based on digitally-reconstructed-fluoroscopy. Our approach may be useful in potentially reducing treatment margins to optimize lung SBRT outcomes. R01-184173.

60 APPLIED LIFE SCIENCES↗

Towards Precision Photometric Type ia Supernova Cosmology With Machine Learning

The revolutionary discovery of dark energy and accelerating cosmic expansion was made with just 42 type Ia supernovae (SNe Ia) in 1999. Since then, large synoptic surveys, e.g., Dark Energy Survey (DES), have observed thousands more SNe Ia and the upcoming Rubin Legacy Survey of Space and Time (LSST) and Roman Space Telescope promise to deliver millions in the next decade. This unprecedented data volume can produce the required precision to unambiguously test concordance cosmology which could represent a monumental shift in our understanding of dark energy and its role in cosmic history. However, extracting a pure SN Ia sample with accurate redshifts for such a large dataset will be a challenge. Specifically, spectroscopic classification will not be possible for the vast majority of discovered objects, and only ~25% will have spectroscopic redshifts. This thesis presents a series of observational and methodological studies designed to address the questions associated with this new era of photometric SN Ia cosmology. First, we present a machine learning method for photometric classification of SNe, Supernova Classification with a COnvolutional Neural Network (SCONE). Photometric classification enables SNe with no spectroscopic information to be confidently categorized, and is a critical component of current and future analysis pipelines. SCONE achieves >99% accuracy distinguishing simulated SNe Ia from non-Ia SNe, and has been integrated into DES, LSST, and Roman analysis pipelines. We also demonstrate the efficacy of SCONE on early-time photometric classification, which will be vital for optimal allocation of spectroscopic resources. We show that SCONE can distinguish between 6 SN types with 75% accuracy on the night of initial discovery, comparable to results in the literature for full-phase SNe. Next, we study current methods for estimating SN Ia redshifts and propose a machine learning alternative that uses SN photometry alone to extract redshift information. Most SNe Ia inherit redshift information from their host galaxy, but the process of matching SNe to the correct host galaxy can be challenging. We systematically analyze the impact of incorrect redshifts from host galaxy mismatch on 5 years of DES SN data, and conclude that improved host matching or redshift estimation methods can reduce our systematic errors by ~10%. In response to this finding, we present a SN photometry-only method for estimating redshifts independent of host galaxy information, Photo-zSNthesis. We show that Photo-zSNthesis redshift estimates are accurate to within 2% across the full redshift range of LSST, a first in the literature. Finally, we focus on the robustness of machine learning (ML) algorithms for real-world and scientific applications. ML models generalize poorly beyond their training set and often experience severe performance degradation when deployed on new data. We demonstrate a general method for improving robustness that achieves new state-of-the-art results on astronomical object classification, wildlife identification, and tumor detection.

79 ASTRONOMY AND ASTROPHYSICS↗

Assimilating partial observation to enhance feedback control of stochastic dynamical systems

Here, in this paper, we present a novel methodology to tackle feedback optimal control problems in scenarios where the exact state of the controlled process is unknown. It integrates data assimilation techniques and optimal control solvers to manage partial observation of the state process, a common occurrence in practical scenarios. Traditional stochastic optimal control methods assume full state observation, which is often not feasible in real-world fluid dynamics control problems. Our approach underscores the significance of utilizing observational data to inform control policy design. Specifically, we introduce a kernel learning backward stochastic differential equation (SDE) filter to enhance data assimilation efficiency and propose a sample-wise stochastic optimization method within the stochastic maximum principle framework. We demonstrate the efficacy and accuracy of our method in the control of advection-diffusion-reaction flow problem and the Dubins airplane maneuvering problem with model uncertainty.

data driven↗

SIFT-based dense pixel tracking on 0.35 T cine-MR images acquired during image-guided radiation therapy with application to gating optimization

Purpose: To first demonstrate the viability of applying an image processing technique for tracking regions on low-contrast cine-MR images acquired during image-guided radiation therapy, and then outline a scheme that uses tracking data for optimizing gating results in a patient-specific manner. Methods: A first-generation MR-IGRT system—treating patients since January 2014—integrates a 0.35 T MR scanner into an annular gantry consisting of three independent Co-60 sources. Obtaining adequate frame rates for capturing relevant patient motion across large fields-of-view currently requires coarse in-plane spatial resolution. This study initially (1) investigate the feasibility of rapidly tracking dense pixel correspondences across single, sagittal plane images (with both moderate signal-to-noise and spatial resolution) using a matching objective for highly descriptive vectors called scale-invariant feature transform (SIFT) descriptors associated to all pixels that describe intensity gradients in local regions around each pixel. To more accurately track features, (2) harmonic analysis was then applied to all pixel trajectories within a region-of-interest across a short training period. In particular, the procedure adjusts the motion of outlying trajectories whose relative spectral power within a frequency bandwidth consistent with respiration (or another form of periodic motion) does not exceed a threshold value that is manually specified following the training period. To evaluate the tracking reliability after applying this correction, conventional metrics—including Dice similarity coefficients (DSCs), mean tracking errors (MTEs), and Hausdorff distances (HD)—were used to compare target segmentations obtained via tracking to manually delineated segmentations. Upon confirming the viability of this descriptor-based procedure for reliably tracking features, the study (3) outlines a scheme for optimizing gating parameters—including relative target position and a tolerable margin about this position—derived from a probability density function that is constructed using tracking results obtained just prior to treatment. Results: The feasibility of applying the matching objective for SIFT descriptors toward pixel-by-pixel tracking on cine-MR acquisitions was first retrospectively demonstrated for 19 treatments (spanning various sites). Both with and without motion correction based on harmonic analysis, sub-pixel MTEs were obtained. A mean DSC value spanning all patients of 0.916 ± 0.001 was obtained without motion correction, with DSC values exceeding 0.85 for all patients considered. While most patients show accurate tracking without motion correction, harmonic analysis does yield substantial gain in accuracy (defined using HDs) for three particularly challenging subjects. An application of tracking toward a gating optimization procedure was then demonstrated that should allow a physician to balance beam-on time and tissue sparing in a patient-specific manner by tuning several intuitive parameters. Conclusions: Tracking results show high fidelity in assessing intrafractional motion observed on cine-MR acquisitions. Incorporating harmonic analysis during a training period improves the robustness of the tracking for challenging targets. The concomitant gating optimization procedure should allow for physicians to quantitatively assess gating effectiveness quickly just prior to treatment in a patient-specific manner.

61 RADIATION PROTECTION AND DOSIMETRY↗

Improved Accuracy for Trotter Simulations Using Chebyshev Interpolation

Quantum metrology allows for measuring properties of a quantum system at the optimal Heisenberg limit. However, when the relevant quantum states are prepared using digital Hamiltonian simulation, the accrued algorithmic errors will cause deviations from this fundamental limit. In this work, we show how algorithmic errors due to Trotterized time evolution can be mitigated through the use of standard polynomial interpolation techniques. Our approach is to extrapolate to zero Trotter step size, akin to zero-noise extrapolation techniques for mitigating hardware errors. We perform a rigorous error analysis of the interpolation approach for estimating eigenvalues and time-evolved expectation values, and show that the Heisenberg limit is achieved up to polylogarithmic factors in the error. Our work suggests that accuracies approaching those of state-of-the-art simulation algorithms may be achieved using Trotter and classical resources alone for a number of relevant algorithmic tasks.

Rendon, Gumaro↗

Dose domain regularization of MLC leaf patterns for highly complex IMRT plans

Purpose: The advent of automated beam orientation and fluence optimization enables more complex intensity modulated radiation therapy (IMRT) planning using an increasing number of fields to exploit the expanded solution space. This has created a challenge in converting complex fluences to robust multileaf collimator (MLC) segments for delivery. A novel method to regularize the fluence map and simplify MLC segments is introduced to maximize delivery efficiency, accuracy, and plan quality. Methods: In this work, we implemented a novel approach to regularize optimized fluences in the dose domain. The treatment planning problem was formulated in an optimization framework to minimize the segmentation-induced dose distribution degradation subject to a total variation regularization to encourage piecewise smoothness in fluence maps. The optimization problem was solved using a first-order primal-dual algorithm known as the Chambolle-Pock algorithm. Plans for 2 GBM, 2 head and neck, and 2 lung patients were created using 20 automatically selected and optimized noncoplanar beams. The fluence was first regularized using Chambolle-Pock and then stratified into equal steps, and the MLC segments were calculated using a previously described level reducing method. Isolated apertures with sizes smaller than preset thresholds of 1–3 bixels, which are square units of an IMRT fluence map from MLC discretization, were removed from the MLC segments. Performance of the dose domain regularized (DDR) fluences was compared to direct stratification and direct MLC segmentation (DMS) of the fluences using level reduction without dose domain fluence regularization. Results: For all six cases, the DDR method increased the average planning target volume dose homogeneity (D95/D5) from 0.814 to 0.878 while maintaining equivalent dose to organs at risk (OARs). Regularized fluences were more robust to MLC sequencing, particularly to the stratification and small aperture removal. The maximum and mean aperture sizes using the DDR were consistently larger than those from DMS for all tested number of segments. Conclusions: The fluence map to MLC segmentation conversion problem was formulated as a secondary optimization problem in the dose domain to minimize the smoothness-regularized dose discrepancy. The large scale optimization problem was solved using a primal-dual algorithm that transformed complicated fluences into maps that were more robust to the MLC segmentation and sequencing, affording fewer and larger segments with minimal degradation to dose distribution.

60 APPLIED LIFE SCIENCES↗

SU-D-BRA-04: Computerized Framework for Marker-Less Localization of Anatomical Feature Points in Range Images Based On Differential Geometry Features for Image-Guided Radiation Therapy

Purpose: To propose a computerized framework for localization of anatomical feature points on the patient surface in infrared-ray based range images by using differential geometry (curvature) features. Methods: The general concept was to reconstruct the patient surface by using a mathematical modeling technique for the computation of differential geometry features that characterize the local shapes of the patient surfaces. A region of interest (ROI) was firstly extracted based on a template matching technique applied on amplitude (grayscale) images. The extracted ROI was preprocessed for reducing temporal and spatial noises by using Kalman and bilateral filters, respectively. Next, a smooth patient surface was reconstructed by using a non-uniform rational basis spline (NURBS) model. Finally, differential geometry features, i.e. the shape index and curvedness features were computed for localizing the anatomical feature points. The proposed framework was trained for optimizing shape index and curvedness thresholds and tested on range images of an anthropomorphic head phantom. The range images were acquired by an infrared ray-based time-of-flight (TOF) camera. The localization accuracy was evaluated by measuring the mean of minimum Euclidean distances (MMED) between reference (ground truth) points and the feature points localized by the proposed framework. The evaluation was performed for points localized on convex regions (e.g. apex of nose) and concave regions (e.g. nasofacial sulcus). Results: The proposed framework has localized anatomical feature points on convex and concave anatomical landmarks with MMEDs of 1.91±0.50 mm and 3.70±0.92 mm, respectively. A statistically significant difference was obtained between the feature points on the convex and concave regions (P<0.001). Conclusion: Our study has shown the feasibility of differential geometry features for localization of anatomical feature points on the patient surface in range images. The proposed framework might be useful for tasks involving feature-based image registration in range-image guided radiation therapy.

60 APPLIED LIFE SCIENCES↗

An optimal guarding scheme for thermal conductivity measurement using a guarded cut-bar technique, part 1 experimental study

In the guarded cut-bar technique, a guard surrounding the measured sample and reference (meter) bars is temperature controlled to carefully regulate heat losses from the sample and reference bars. Guarding is typically carried out by matching the temperature profiles between the guard and the test stack of sample and meter bars. Problems arise in matching the profiles, especially when the thermal conductivitiesof the meter bars and of the sample differ, as is usually the case. In a previous numerical study, the applied guarding condition (guard temperature profile) was found to be an important factor in measurement accuracy. Different from the linear-matched or isothermal schemes recommended in literature, the optimal guarding condition is dependent on the system geometry and thermal conductivity ratio of sample to meter bar. To validate the numerical results, an experimental study was performed to investigate the resulting error under different guarding conditions using stainless steel 304 as both the sample and meter bars. The optimal guarding condition was further verified on a certified reference material, pyroceram 9606, and 99.95% pure iron whose thermal conductivities are much smaller and much larger, respectively, than that of the stainless steel meter bars. Additionally, measurements are performed using three different inert gases to show the effect of the insulation effective thermal conductivity on measurement error, revealing low conductivity, argon gas, gives the lowest error sensitivity when deviating from the optimal condition. The result of this study provides a general guideline for the specific measurement method and for methods requiring optimal guarding or insulation.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Machine learning-enhanced model-based scenario optimization for DIII-D

Abstract Scenario development in tokamaks is an open area of investigation that can be approached in a variety of different ways. Experimental trial and error has been the traditional method, but this required a massive amount of experimental time and resources. As high fidelity predictive models have become available, offline development and testing of proposed scenarios has become an option to reduce the required experimental resources. The use of predictive models also offers the possibility of using a numerical optimization process to find the controllable inputs that most closely achieve the desired plasma state. However, this type of optimization can require as many as hundreds or thousands of predictive simulation cases to converge to a solution; many of the commonly used high fidelity models have high computational burdens, so it is only reasonable to run a handful of predictive simulations. In order to make use of numerical optimization approaches, a compromise needs to be found between model fidelity and computational burden. This compromise can be achieved using neural networks surrogates of high fidelity models that retain nearly the same level of accuracy as the models they are trained to replicate while reducing the computation time by orders of magnitude. In this work, a model-based numerical optimization tool for scenario development is described. The predictive model used by the optimizer includes neural network surrogate models integrated into the fast Control-Oriented Transport simulation framework. This optimization scheme is able to converge to the optimal values of the controllable inputs that produce the target plasma scenario by running thousands of predictive simulations in under an hour without sacrificing too much prediction accuracy.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Acoustic inverse scattering via Helmholtz operator factorization and optimization

We present a joint acoustic/seismic inverse scattering and finite-frequency (reflection) tomography program, formulated as a coupled set of optimization problems, in terms of inhomogeneous Helmholtz equations. We use a higher order finite difference scheme for these Helmholtz equations to guarantee sufficient accuracy. We adapt a structured approximate direct solver for the relevant systems of algebraic equations, which addresses storage requirements through compression, to yield a complexity for computing the gradients or images in the optimization problems that consists of two parts, viz., the cost for all the matrix factorizations which is roughly O(rN) (for example O(rNlogN) when d = 2) times the number of frequencies, and the cost for all solutions by substitution which is roughly O(N) (for example O(Nlog(rlogN)) when d = 2) times the number of frequencies times the number of sources (events), where N = n{sup d} if n is the number of grid samples in any direction, and r is a parameter depending on the preset accuracy and the problem at hand. With this complexity, the multi-frequency approach to inverse scattering and finite-frequency tomography becomes computationally feasible with large data sets, in dimensions d = 2 and 3.

97 MATHEMATICS AND COMPUTING↗

Communication: An adaptive configuration interaction approach for strongly correlated electrons with tunable accuracy

We introduce a new procedure for iterative selection of determinant spaces capable of describing highly correlated systems. This adaptive configuration interaction (ACI) determines an optimal basis by an iterative procedure in which the determinant space is expanded and coarse grained until self-consistency. Two importance criteria control the selection process and tune the ACI to a user-defined level of accuracy. The ACI is shown to yield potential energy curves of N{sub 2} with nearly constant errors, and it predicts singlet-triplet splittings of acenes up to decacene that are in good agreement with the density matrix renormalization group.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Performance and power modeling and prediction using MuMMI and 10 machine learning methods

Energy-efficient scientific applications require insight into how high performance computing system features impact the applications' power and performance. This insight can result from the development of performance and power models. Here, in this article, we use the modeling and prediction tool MuMMI (Multiple Metrics Modeling Infrastructure) and 10 machine learning methods to model and predict performance and power consumption and compare their prediction error rates. We use an algorithm-based fault-tolerant linear algebra code and a multilevel checkpointing fault-tolerant heat distribution code to conduct our modeling and prediction study on the Cray XC40 Theta and IBM BG/Q Mira at Argonne National Laboratory and the Intel Haswell cluster Shepard at Sandia National Laboratories. Our experimental results show that the prediction error rates in performance and power using MuMMI are less than 10% for most cases. By utilizing the models for runtime, node power, CPU power, and memory power, we identify the most significant performance counters for potential application optimizations, and we predict theoretical outcomes of the optimizations. Based on two collected datasets, we analyze and compare the prediction accuracy in performance and power consumption using MuMMI and 10 machine learning methods.

97 MATHEMATICS AND COMPUTING↗

Co-Optimization of CO 2 -EOR and Storage Processes in Mature Oil Reservoirs

This article presents an optimization methodology for CO 2 enhanced oil recovery in partially depleted reservoirs. A field-scale compositional reservoir flow model was developed for assessing the performance history of an active CO 2 flood and for optimizing both oil production and CO 2 storage in the Farnsworth Unit (FWU), Ochiltree County, Texas. A geological framework model constructed from geophysical, geological, and engineering data acquired from the FWU was the basis for all reservoir simulations and the optimization method. An equation of state was calibrated with laboratory fluid analyses and subsequently used to predict the thermodynamic minimum miscible pressure (MMP). Initial history calibrations of primary, secondary and tertiary recovery were conducted as the basis for the study. After a good match was achieved, an optimization approach consisting of a proxy or surrogate model was constructed with a polynomial response surface method (PRSM). The PRSM utilized an objective function that maximized both oil recovery and CO 2 storage. Experimental design was used to link uncertain parameters to the objective function. Control variables considered in this study included: water alternating gas cycle and ratio, production rates and bottom-hole pressure of injectors and producers. Other key parameters considered in the modeling process were CO 2 purchase, gas recycle and addition of infill wells and/or patterns. The PRSM proxy model was ‘trained’ or calibrated with a series of training simulations. This involved an iterative process until the surrogate model reached a specific validation criterion. A sensitivity analysis was first conducted to ascertain which of these control variables to retain in the surrogate model. A genetic algorithm with a mixed-integer capability optimization approach was employed to determine the optimum developmental strategy to maximize both oil recovery and CO 2 storage. The proxy model reduced the computational cost significantly. The validation criteria of the reduced order model ensured accuracy in the dynamic modeling results. The prediction outcome suggested robustness and reliability of the genetic algorithm for optimizing both oil recovery and CO 2 storage. The reservoir modeling approach used in this study illustrates an improved approach to optimizing oil production and CO 2 storage within partially depleted oil reservoirs such as FWU. Lastly, this study may serve as a benchmark for potential CO 2 –EOR projects in the Anadarko basin and/or geologically similar basins throughout the world.

02 PETROLEUM↗

A Chaos Theoretic Approach to Animal Activity Recognition

Animal activity is a descriptor that can be potentially used to track the health and well-being, which is obviously very important to improve the management process and productivity of farms. This paper deals with an animal behavior recognition problem using a chaos theory approach where we adopt such a technique for automatic classification of calves behavioral states. Two main mutually exclusive behaviors are of interest, namely, lying and standing/walking with six possible activities: feeding, drinking water, drinking milk, playing, rumination, and neutral. The time series generated by ear-tags with a 3D-accelerometer after a wavelet denoising transformation and frequency stabilization are treated as representations of the nonlinear dynamical system. The dynamical system of a certain animal state exhibits specific strange attractor in a phase space. A characterization of such an attractor is performed through metric, dynamic, and topological invariants including Lyapunov exponent, correlation dimension, length of a phase trajectory, sum of edges forming a convex hull and others. These measures are used as a feature vector for the subsequent classification. In a cross validation scheme, six classifiers are built on each training set, and the hyper-parameters are optimized using an inner validation set. The classifier that reaches the highest accuracy on the inner validation set is used to classify the outer validation set. It is shown that this approach can be useful at predicting activity states as an alternative methodology for the animal behavior state recognition problem with acceptable classification accuracy. Furthermore it is possible to include this procedure as part of an ensemble method in machine learning where a combination of different models is used.

97 MATHEMATICS AND COMPUTING↗

An efficient algorithm for the projection of a point on the intersection of two hyperplanes and a box in $\mathbb {R}^n$

Here, we present an efficient strongly polynomial algorithm for the projection of a point on the intersection of two hyperplanes and a box in \(\mathbb {R}^n\) . Interior point methods are the most efficient algorithm in the literature to solve this problem. While efficient in practice, the complexity of interior-point methods is bounded by a polynomial in the dimension of the problem and in the accuracy of the solution. Moreover, their efficiency is highly dependent on a series of parameters depending on the specific method chosen (especially for nonlinear problems), such as step size, barrier parameter, accuracy, among others. We propose a new method based on the KKT optimality conditions. In this method, we write the problem as a function of the Lagrangian multipliers of the hyperplanes and seek to find the pair of multipliers that corresponds to the optimal solution. We prove that the algorithm has complexity \(O(n^2 \log n)\) .

42 ENGINEERING↗

Effect of particle size and moisture on flow performance of loblolly pine anatomical fractions: Experimental findings and model predictions

The rising energy demand has highlighted biomass as a promising next-generation energy source. However, commercializing biomass-derived energy faces challenges, particularly in handling biomass feedstock. Factors like particle size, shape, moisture content, and surface roughness significantly impact biomass flowability. This study addresses a crucial knowledge gap by examining the effects of particle size and moisture content on the flow behavior and shear properties of different anatomical fractions of loblolly pine (Pinus taeda). The bulk shear behavior was examined using a Schulze ring shear tester, while flow performance was tested through gravity-driven flow experiments in a variable wedge-shape hopper. Results were incorporated into empirical and machine learning-based flow prediction models to evaluate their accuracy and limitations. The study found that samples with higher moisture content show higher unconfined yield strength. The critical arching distance increased with particle size, e.g., from approximately 13 and 33 mm for 2- and 6-mm whole chips, respectively at a 32-degree inclination angle. Conversely, the flow rate decreased for a given hopper opening as particle size increased. For instance, at a 60-mm hopper opening and a 32-degree inclination angle, the mass flow rates for 2- and 6-mm whole chips were 7.83 and 6.42 tonne/h, respectively. The empirical model consistently overpredicted the mass flow rate for all anatomical fractions, while the machine learning model more accurately predicted the central tendency of flow rate but was insensitive to varying tissue proportions. These novel findings provide comprehensive characterization of anatomical fractions, reveal significant combined effects of particle size and moisture content on biomass flow behavior, and demonstrate a better predictive accuracy of a machine learning model, all of which are useful for optimizing material handling strategies and biomass utilization technologies in the industry.

09 - BIOMASS FUELS↗

hp -VPINNs: Variational physics-informed neural networks with domain decomposition

We formulate a general framework for hp-variational physics-informed neural networks (hp-VPINNs) based on the nonlinear approximation of shallow and deep neural networks and hp-refinement via domain decomposition and projection onto the space of high-order polynomials. The trial space is the space of neural network, which is defined globally over the entire computational domain, while the test space contains piecewise polynomials. Specifically in this study, the hp-refinement corresponds to a global approximation with a local learning algorithm that can efficiently localize the network parameter optimization. Here, we demonstrate the advantages of hp-VPINNs in both accuracy and training cost for several numerical examples of function approximation and in solving differential equations.

42 ENGINEERING↗