Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “sampling algorithm”

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 1,495 records · Page 83

Adaptation of the CVT algorithm for catheter optimization in high dose rate brachytherapy

Purpose: An innovative, simple, and fast method to optimize the number and position of catheters is presented for prostate and breast high dose rate (HDR) brachytherapy, both for arbitrary templates or template-free implants (such as robotic templates).Methods: Eight clinical cases were chosen randomly from a bank of patients, previously treated in our clinic to test our method. The 2D Centroidal Voronoi Tessellations (CVT) algorithm was adapted to distribute catheters uniformly in space, within the maximum external contour of the planning target volume. The catheters optimization procedure includes the inverse planning simulated annealing algorithm (IPSA). Complete treatment plans can then be generated from the algorithm for different number of catheters. The best plan is chosen from different dosimetry criteria and will automatically provide the number of catheters and their positions. After the CVT algorithm parameters were optimized for speed and dosimetric results, it was validated against prostate clinical cases, using clinically relevant dose parameters. The robustness to implantation error was also evaluated. Finally, the efficiency of the method was tested in breast interstitial HDR brachytherapy cases.Results: The effect of the number and locations of the catheters on prostate cancer patients was studied. Treatment plans with a better or equivalent dose distributions could be obtained with fewer catheters. A better or equal prostate V100 was obtained down to 12 catheters. Plans with nine or less catheters would not be clinically acceptable in terms of prostate V100 and D90. Implantation errors up to 3 mm were acceptable since no statistical difference was found when compared to 0 mm error (p > 0.05). No significant difference in dosimetric indices was observed for the different combination of parameters within the CVT algorithm. A linear relation was found between the number of random points and the optimization time of the CVT algorithm. Because the computation time decrease with the number of points and that no effects were observed on the dosimetric indices when varying the number of sampling points and the number of iterations, they were respectively fixed to 2500 and to 100. The computation time to obtain ten complete treatments plans ranging from 9 to 18 catheters, with the corresponding dosimetric indices, was 90 s. However, 93% of the computation time is used by a research version of IPSA. For the breast, on average, the Radiation Therapy Oncology Group recommendations would be satisfied down to 12 catheters. Plans with nine or less catheters would not be clinically acceptable in terms of V100, dose homogeneity index, and D90.Conclusions: The authors have devised a simple, fast and efficient method to optimize the number and position of catheters in interstitial HDR brachytherapy. The method was shown to be robust for both prostate and breast HDR brachytherapy. More importantly, the computation time of the algorithm is acceptable for clinical use. Ultimately, this catheter optimization algorithm could be coupled with a 3D ultrasound system to allow real-time guidance and planning in HDR brachytherapy.

61 RADIATION PROTECTION AND DOSIMETRY↗

An adaptive Hessian approximated stochastic gradient MCMC method

Bayesian approaches have been successfully integrated into training deep neural networks. One popular family is stochastic gradient Markov chain Monte Carlo methods (SG-MCMC), which have gained increasing interest due to their ability to handle large datasets and the potential to avoid overfitting. Although standard SG-MCMC methods have shown great performance in a variety of problems, they may be inefficient when the random variables in the target posterior densities have scale differences or are highly correlated. Here, we present an adaptive Hessian approximated stochastic gradient MCMC method to incorporate local geometric information while sampling from the posterior. The idea is to apply stochastic approximation (SA) to sequentially update a preconditioning matrix at each iteration. The preconditioner possesses second-order information and can guide the random walk of a sampler efficiently. Instead of computing and saving the full Hessian of the log posterior, we use limited memory of the samples and their stochastic gradients to approximate the inverse Hessian-vector multiplication in the updating formula. Moreover, by smoothly optimizing the preconditioning matrix via SA, our proposed algorithm can asymptotically converge to the target distribution with a controllable bias under mild conditions. To reduce the training and testing computational burden, we adopt a magnitude-based weight pruning method to enforce the sparsity of the network. Our method is user-friendly and demonstrates better learning results compared to standard SG-MCMC updating rules. The approximation of inverse Hessian alleviates storage and computational complexities for large dimensional models. Numerical experiments are performed on several problems, including sampling from 2D correlated distribution, synthetic regression problems, and learning the numerical solutions of heterogeneous elliptic PDE. The numerical results demonstrate great improvement in both the convergence rate and accuracy.

97 MATHEMATICS AND COMPUTING↗

An Adaptive Sparse-Grid High-Order Stochastic Collocation Method for Bayesian Inference in Groundwater Reactive Transport Modeling

Although Bayesian analysis has become vital to the quantification of prediction uncertainty in groundwater modeling, its application has been hindered due to the computational cost associated with numerous model executions needed for exploring the posterior probability density function (PPDF) of model parameters. This is particularly the case when the PPDF is estimated using Markov Chain Monte Carlo (MCMC) sampling. In this study, we develop a new approach that improves computational efficiency of Bayesian inference by constructing a surrogate system based on an adaptive sparse-grid high-order stochastic collocation (aSG-hSC) method. Unlike previous works using first-order hierarchical basis, we utilize a compactly supported higher-order hierar- chical basis to construct the surrogate system, resulting in a significant reduction in the number of computational simulations required. In addition, we use hierarchical surplus as an error indi- cator to determine adaptive sparse grids. This allows local refinement in the uncertain domain and/or anisotropic detection with respect to the random model parameters, which further improves computational efficiency. Finally, we incorporate a global optimization technique and propose an iterative algorithm for building the surrogate system for the PPDF with multiple significant modes. Once the surrogate system is determined, the PPDF can be evaluated by sampling the surrogate system directly with very little computational cost. The developed method is evaluated first using a simple analytical density function with multiple modes and then using two synthetic groundwater reactive transport models. The groundwater models represent different levels of complexity; the first example involves coupled linear reactions and the second example simulates nonlinear ura- nium surface complexation. The results show that the aSG-hSC is an effective and efficient tool for Bayesian inference in groundwater modeling in comparison with conventional MCMC sim- ulations. The computational efficiency is expected to be more beneficial to more computational expensive groundwater problems.

Zhang, Guannan↗

Partitioned Quantum Subspace Expansion

We present an iterative generalisation of the quantum subspace expansion algorithm used with a Krylov basis. The iterative construction connects a sequence of subspaces via their lowest energy states. Diagonalising a Hamiltonian in a given Krylov subspace requires the same quantum resources in both the single step and sequential cases. We propose a variance-based criterion for determining a good iterative sequence and provide numerical evidence that these good sequences display improved numerical stability over a single step in the presence of finite sampling noise. Implementing the generalisation requires additional classical processing with a polynomial overhead in the subspace dimension. By exchanging quantum circuit depth for additional measurements the quantum subspace expansion algorithm appears to be an approach suited to near term or early error-corrected quantum hardware. Our work suggests that the numerical instability limiting the accuracy of this approach can be substantially alleviated in a parameter-free way.

97 MATHEMATICS AND COMPUTING↗

Advances in machine-learning-based sampling motivated by lattice quantum chromodynamics

Sampling from known probability distributions is a ubiquitous task in computational science, underlying calculations in domains from linguistics to biology and physics. Generative machine-learning (ML) models have emerged as a promising tool in this space, building on the success of this approach in applications such as image, text, and audio generation. Often, however, generative tasks in scientific domains have unique structures and features—such as complex symmetries and the requirement of exactness guarantees—that present both challenges and opportunities for ML. This Perspective outlines the advances in ML-based sampling motivated by lattice quantum field theory, in particular for the theory of quantum chromodynamics. Enabling calculations of the structure and interactions of matter from our most fundamental understanding of particle physics, lattice quantum chromodynamics is one of the main consumers of open-science supercomputing worldwide. Here, the design of ML algorithms for this application faces profound challenges, including the necessity of scaling custom ML architectures to the largest supercomputers, but also promises immense benefits, and is spurring a wave of development in ML-based sampling more broadly. In lattice field theory, if this approach can realize its early promise it will be a transformative step towards first-principles physics calculations in particle, nuclear and condensed matter physics that are intractable with traditional approaches.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

MO-G-12A-01: Quantitative Imaging Metrology: What Should Be Assessed and How?

The first two symposia in the Quantitative Imaging Track focused on 1) the introduction of quantitative imaging (QI) challenges and opportunities, and QI efforts of agencies and organizations such as the RSNA, NCI, FDA, and NIST, and 2) the techniques, applications, and challenges of QI, with specific examples from CT, PET/CT, and MR. This third symposium in the QI Track will focus on metrology and its importance in successfully advancing the QI field. While the specific focus will be on QI, many of the concepts presented are more broadly applicable to many areas of medical physics research and applications. As such, the topics discussed should be of interest to medical physicists involved in imaging as well as therapy. The first talk of the session will focus on the introduction to metrology and why it is critically important in QI. The second talk will focus on appropriate methods for technical performance assessment. The third talk will address statistically valid methods for algorithm comparison, a common problem not only in QI but also in other areas of medical physics. The final talk in the session will address strategies for publication of results that will allow statistically valid meta-analyses, which is critical for combining results of individual studies with typically small sample sizes in a manner that can best inform decisions and advance the field. Learning Objectives: Understand the importance of metrology in the QI efforts. Understand appropriate methods for technical performance assessment. Understand methods for comparing algorithms with or without reference data (i.e., “ground truth”). Understand the challenges and importance of reporting results in a manner that allows for statistically valid meta-analyses.

60 APPLIED LIFE SCIENCES↗

Sample Complexity of Power System State Estimation Using Matrix Completion

In this paper, we propose an analytical framework to quantify the amount of data samples needed to obtain accurate state estimation in a power system - a problem known as sample complexity analysis in computer science. Motivated by the increasing adoption of distributed energy resources into the distribution-level grids, it becomes imperative to estimate the state of distribution grids in order to ensure stable operation. Traditional power system state estimation techniques mainly focus on the transmission network which involve solving an overdetermined system and eliminating bad data. However, distribution networks are typically underdetermined due to the large number of connection points and high cost of pervasive installation of measurement devices. In this paper, we consider the recently proposed state-estimation method for underdetermined systems that is based on matrix completion. In particular, a constrained matrix completion algorithm was proposed, wherein the standard matrix completion problem is augmented with additional equality constraints representing the physics (namely power-flow constraints). We analyze the sample complexity of this general method by proving an upper bound on the sample complexity that depends directly on the properties of these constraints that can lower number of needed samples as compared to the unconstrained problem. To demonstrate the improvement that the constraints add to state estimation, we test the method on a 141-bus distribution network case study and compare it to the traditional least squares minimization state estimation method.

24 POWER TRANSMISSION AND DISTRIBUTION↗

The discriminant power of bubble wall velocities: gravitational waves and electroweak baryogenesis

A precise determination of the bubble wall velocity v$_{w}$ is crucial for making accurate predictions of the baryon asymmetry and gravitational wave (GW) signals in models of electroweak baryogenesis (EWBG). Working in the local thermal equilibrium approximation, we exploit entropy conservation to present efficient algorithms for computing v$_{w}$, significantly streamlining the calculation. We then explore the parameter dependencies of v$_{w}$, focusing on two sample models capable of enabling a strong first-order electroweak phase transition: a ℤ$_{2}$-symmetric singlet extension of the SM, and a model for baryogenesis with CP violation in the dark sector. We study correlations among v$_{w}$ and the two common measures of phase transition strength, α$_{n}$ and v$_{n}$/T$_{n}$. Interestingly, we find a relatively model-insensitive relationship between v$_{n}$/T$_{n}$ and α$_{n}$. We also observe an upper bound on α$_{n}$ for the deflagration/hybrid wall profiles naturally compatible with EWBG, the exact value for which varies between models, significantly impacting the strength of the GW signals. In summary, our work provides a framework for exploring the feasibility of EWBG models in light of future GW signals.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Accelerating Machine Learning Inference with GPUs in ProtoDUNE Data Processing

Abstract We study the performance of a cloud-based GPU-accelerated inference server to speed up event reconstruction in neutrino data batch jobs. Using detector data from the ProtoDUNE experiment and employing the standard DUNE grid job submission tools, we attempt to reprocess the data by running several thousand concurrent grid jobs, a rate we expect to be typical of current and future neutrino physics experiments. We process most of the dataset with the GPU version of our processing algorithm and the remainder with the CPU version for timing comparisons. We find that a 100-GPU cloud-based server is able to easily meet the processing demand, and that using the GPU version of the event processing algorithm is two times faster than processing these data with the CPU version when comparing to the newest CPUs in our sample. The amount of data transferred to the inference server during the GPU runs can overwhelm even the highest-bandwidth network switches, however, unless care is taken to observe network facility limits or otherwise distribute the jobs to multiple sites. We discuss the lessons learned from this processing campaign and several avenues for future improvements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

A rigorous sequential update strategy for parallel kinetic Monte Carlo simulation

The kinetic Monte Carlo (kMC) method is used in many scientific fields in applications involving rare-event transitions. Due to its discrete stochastic nature, efforts to parallelize kMC approaches often produce unbalanced time evolutions requiring complex implementations to ensure correct statistics. In the context of parallel kMC, the sequential update technique has shown promise by generating high quality distributions with high relative efficiencies for short-range systems. In this work, we provide an extension of the sequential update method in a parallel context that rigorously obeys detailed balance, which guarantees exact equilibrium statistics for all parallelization settings. Furthermore, our approach also preserves nonequilibrium dynamics with minimal error for many parallelization settings, and can be used to achieve highly precise sampling.

36 MATERIALS SCIENCE↗

Approximate Bayesian Model Inversion for PDEs with Heterogeneous and State-Dependent Coefficients

We present two approximate Bayesian inference methods for parameter estimation in partial differential equation (PDE) models with space-dependent and state-dependent parameters. We demonstrate that these methods provide accurate and cost-effective alternatives to Markov Chain Monte Carlo simulation. We assume a parameterized Gaussian prior on the unknown functions, and approximate the posterior density by a parameterized multivariate Gaussian density. The parameters of the prior and posterior are estimated from sparse observations of the PDE model's states and the unknown functions themselves by maximizing the evidence lower bound (ELBO), a lower bound on the log marginal likelihood of the observations. The first method, Laplace-EM, employs the expectation maximization algorithm to maximize the ELBO, with a Laplace approximation of the posterior on the E-step, and minimization of a Kullback-Leibler divergence on the M-step. The second method, DSVI-EB, employs the doubly stochastic variational inference (DSVI) algorithm, in which the ELBO is maximized via gradient-based stochastic optimization, with nosiy gradients computed via simple Monte Carlo sampling and Gaussian backpropagation. We apply these methods to identifying diffusion coefficients in linear and nonlinear diffusion equations, and we find that both methods provide accurate estimates of posterior densities and the hyperparameters of Gaussian priors. While the Laplace-EM method is more accurate, it requires computing Hessians of the physics model. The DSVI-EB method is found to be less accurate but only requires gradients of the physics model.

approximate Bayesian inference, model inversion, v↗

An isotope identification injection study with GammaTracker

GammaTracker is a portable handheld radioisotope identifier using position sensitive CdZnTe detectors. High confidence isotope identification is possible on GammaTracker owing to the system's relatively high energy resolution and count rate sensitivity. A study was undertaken to evaluate the isotope identification performance of a prototype unit. Background and source spectra for various nuclides were measured and then randomly sampled to simulate various integration times and source intensities. The resulting spectral data sets were then run through the isotope identification algorithm to determine the probability of detection and the false alarm rate for each nuclide. The process was repeated for various isotope identification input parameters until an optimized set was achieved. This paper presents results from the injection study.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Accounting for scattering and reabsorption in the analysis of luminescence spectra of nanoparticles

An algorithm and a program are developed allowing the correction of experimentally measured luminescence spectra of nanocomposite materials with the aim to determine the luminescence spectra of the nanoparticles or of the local regions of the sample. The optical characteristics of CdS nanoparticles synthesised in a high-pressure polyethylene and in the orthoxylene solution are studied. It is shown that the luminescence spectra of the nanoparticles differ from those experimentally observed due to the scattering and reabsorption in the samples. In the region of wavelengths 500 - 550 nm a ghost band may arise due to the scattering and reabsorption of the luminescence in the sample near the edge of the CdS absorption band. (optical technologies in biophysics and medicine)

36 MATERIALS SCIENCE↗

3D imaging of magnetic domains in Nd 2 Fe 14 B using scanning hard X-ray nanotomography

Nanoscale structural and electronic heterogeneities are prevalent in condensed matter physics. Investigating these heterogeneities in 3D has become an important task for understanding material properties. To provide a tool to unravel the connection between nanoscale heterogeneity and macroscopic emergent properties in magnetic materials, scanning transmission X-ray microscopy (STXM) is combined with X-ray magnetic circular dichroism. A vector tomography algorithm has been developed to reconstruct the full 3D magnetic vector field without any prior noise assumptions or knowledge about the sample. Two tomographic scans around the vertical axis are acquired on single-crystalline Nd 2 Fe 14 B pillars tilted at two different angles, with 2D STXM projections recorded using a focused 120 nm X-ray beam with left and right circular polarization. Image alignment and iterative registration have been implemented based on the 2D STXM projections for the two tilts. Dichroic projections obtained from difference images are used for the tomographic reconstruction to obtain the 3D magnetization distribution at the nanoscale.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Simulated Contrast of Two Dislocations

A three-dimensional Monte Carlo simulation algorithm is used to study the contrast of two dislocations perpendicular to the irradiated surface of an n-doped silicon sample in the electron beam induced current mode. The dislocations are positioned in the irradiation trajectory, and each of both is considered as a cylinder where the minority carrier diffusion length varies abruptly from a low inside dislocations up to a high value outside dislocations. The EBIC contrast was obtained by simulating the random diffusion of carriers generated at point-like sources randomly distributed within the generation volume. Results are analyzed on the basis of change in the generation volume in the bulk of the sample and of carrier trapping process inside dislocations. The EBIC contrast increases with the increase of the electron beam energy. It also increases when the minority diffusion length inside dislocations, or their separating distance decreases.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Two-Stage Distributionally Robust Conic Linear Programming over 1-Wasserstein Balls

Here, this paper studies two-stage distributionally robust conic linear programming under constraint uncertainty over type-1 Wasserstein balls. We present optimality conditions for the dual of the worst-case expectation problem, which characterizes worst-case uncertain parameters for its inner maximization problem. This condition offers an alternative proof, a counterexample, and an extension to previous works. Additionally, the condition highlights the potential advantage of a specific distance metric for out-of-sample performance, as exemplified in a numerical study on a facility location problem with demand uncertainty. Furthermore, cutting-plane-based algorithms, equipped with a unified scenario generation framework, are proposed for addressing both unbounded support and second-stage dual feasible regions, with a finite convergence proof under less stringent assumptions.

Wasserstein↗

AnisWave2D

AnisWave2D is a 2D finite-difference code for simulating seismic wave propagation in fully anisotropic materials. The code is implemented to run in parallel over multiple processors and is fully portable. A mesh refinement algorithm has been utilized to allow the grid-spacing to be tailored to the velocity model, avoiding the over-sampling of high-velocity materials that usually occurs in fixed-grid schemes.

Toomey, Aoife↗

Lead Slowing Down Spectrometry Analysis of Data from Measurements on Nuclear Fuel

Improved non-destructive assay of isotopic masses in used nuclear fuel would be valuable for nuclear safeguards operations associated with the transport, storage and reprocessing of used nuclear fuel. Our collaboration is examining the feasibility of using lead slowing down spectrometry techniques to assay the isotopic fissile masses in used nuclear fuel assemblies. We present the application of our analysis algorithms on measurements conducted with a lead spectrometer. The measurements involved a single fresh fuel pin and discrete 239Pu and 235U samples. We are able to describe the isotopic fissile masses with root mean square errors over seven different configurations to 6.35% for 239Pu and 2.7% for 235U over seven different configurations. Funding Source(s):

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗