Search NASA⌕ Search

SEARCH · Search NASA

Results for “statistical processing methods”

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 1,747 records · Page 97

Quantum simulations of excited states with active-space downfolded Hamiltonians

Many-body techniques based on the double unitary coupled cluster ansatz (DUCC) can be used to downfold electronic Hamiltonians into low-dimensional active spaces. It can be shown that the resulting dimensionality reduced Hamiltonians are amenable for quantum computing. Recent studies performed for several benchmark systems using quantum phase estimation (QPE) algorithms demonstrated that these formulations can recover a significant portion of ground-state dynamical correlation effects that stem from the electron excitations outside of the active space. These results have also been confirmed in studies of ground-state potential energy surfaces using quantum simulators. In this letter, we study the effectiveness of the DUCC formalism in describing excited states. We also emphasize the role of the QPE formalism and its stochastic nature in discovering/identifying excited states or excited-state processes in situations when the knowledge about the true configurational structure of a sought after excited state is limited or postulated (due to the specific physics driving excited-state processes of interest). In this context, we can view the QPE algorithm as an engine for verifying various hypotheses for excited-state processes and providing statistically meaningful results that correspond to the electronic state(s) with the largest overlap with a postulated configurational structure. We illustrate these ideas on examples of strongly cor- related molecular systems, characterized by small energy gaps and high density of quasi-degenerate states around the Fermi level.

quantum chemistry, quantum computing, computationa↗

ARCHER{sub RT} – A GPU-based and photon-electron coupled Monte Carlo dose computing engine for radiation therapy: Software development and application to helical tomotherapy

Purpose: Using the graphical processing units (GPU) hardware technology, an extremely fast Monte Carlo (MC) code ARCHER{sub RT} is developed for radiation dose calculations in radiation therapy. This paper describes the detailed software development and testing for three clinical TomoTherapy® cases: the prostate, lung, and head and neck. Methods: To obtain clinically relevant dose distributions, phase space files (PSFs) created from optimized radiation therapy treatment plan fluence maps were used as the input to ARCHER{sub RT}. Patient-specific phantoms were constructed from patient CT images. Batch simulations were employed to facilitate the time-consuming task of loading large PSFs, and to improve the estimation of statistical uncertainty. Furthermore, two different Woodcock tracking algorithms were implemented and their relative performance was compared. The dose curves of an Elekta accelerator PSF incident on a homogeneous water phantom were benchmarked against DOSXYZnrc. For each of the treatment cases, dose volume histograms and isodose maps were produced from ARCHER{sub RT} and the general-purpose code, GEANT4. The gamma index analysis was performed to evaluate the similarity of voxel doses obtained from these two codes. The hardware accelerators used in this study are one NVIDIA K20 GPU, one NVIDIA K40 GPU, and six NVIDIA M2090 GPUs. In addition, to make a fairer comparison of the CPU and GPU performance, a multithreaded CPU code was developed using OpenMP and tested on an Intel E5-2620 CPU. Results: For the water phantom, the depth dose curve and dose profiles from ARCHER{sub RT} agree well with DOSXYZnrc. For clinical cases, results from ARCHER{sub RT} are compared with those from GEANT4 and good agreement is observed. Gamma index test is performed for voxels whose dose is greater than 10% of maximum dose. For 2%/2mm criteria, the passing rates for the prostate, lung case, and head and neck cases are 99.7%, 98.5%, and 97.2%, respectively. Due to specific architecture of GPU, modified Woodcock tracking algorithm performed inferior to the original one. ARCHER{sub RT} achieves a fast speed for PSF-based dose calculations. With a single M2090 card, the simulations cost about 60, 50, 80 s for three cases, respectively, with the 1% statistical error in the PTV. Using the latest K40 card, the simulations are 1.7–1.8 times faster. More impressively, six M2090 cards could finish the simulations in 8.9–13.4 s. For comparison, the same simulations on Intel E5-2620 (12 hyperthreading) cost about 500–800 s. Conclusions: ARCHER{sub RT} was developed successfully to perform fast and accurate MC dose calculation for radiotherapy using PSFs and patient CT phantoms.

60 APPLIED LIFE SCIENCES↗

A simulation study exploring the effects of sensor spatial resolution on estimates of cloud cover from satellites.

Investigation of the effects of sensor spatial resolution on estimating the amount of clouds covering the earth by simulating various cloud distributions and sizes, and measuring the known cloud amount with resolution of different sizes using a cloud-no cloud threshold technique often applied in automatic data processing. Cloud amount statistics have been tabulated for a three-orders-of-magnitude range in the ratio (R) of areal cloud size to areal resolution size for seven cloud amounts between 6 and 90%. Three different cloud patterns were used. These were 1) a regularly spaced pattern of homogeneous dots arranged in rows and columns (to simulate cloud streets), 2) a randomly spaced pattern of the same dots (to simulate randomly oriented cumulus clouds), and 3) a heterogeneous cloud size distribution irregularly spaced (to simulate a view of different cloud types and sizes). Two cloud amount estimation techniques were tested. Cloud amounts of 100% (method 1) and 50% (method 2) were assigned to partially filled resolution elements. Using criteria applicable to some studies carried out in the past, it is shown that cloud amount estimations can be in error by as much as 86 and 38%, respectively, for the two methods. Nomograms have been developed which subtantially improve the estimate of the true cloud cover for R less than 100, provided that R can be determined.

Shenk, W. E.↗

Multitaper Spectral Analysis and Wavelet Denoising Applied to Helioseismic Data

Estimates of solar normal mode frequencies from helioseismic observations can be improved by using Multitaper Spectral Analysis (MTSA) to estimate spectra from the time series, then using wavelet denoising of the log spectra. MTSA leads to a power spectrum estimate with reduced variance and better leakage properties than the conventional periodogram. Under the assumption of stationarity and mild regularity conditions, the log multitaper spectrum has a statistical distribution that is approximately Gaussian, so wavelet denoising is asymptotically an optimal method to reduce the noise in the estimated spectra. We find that a single m-upsilon spectrum benefits greatly from MTSA followed by wavelet denoising, and that wavelet denoising by itself can be used to improve m-averaged spectra. We compare estimates using two different 5-taper estimates (Stepian and sine tapers) and the periodogram estimate, for GONG time series at selected angular degrees l. We compare those three spectra with and without wavelet-denoising, both visually, and in terms of the mode parameters estimated from the pre-processed spectra using the GONG peak-fitting algorithm. The two multitaper estimates give equivalent results. The number of modes fitted well by the GONG algorithm is 20% to 60% larger (depending on l and the temporal frequency) when applied to the multitaper estimates than when applied to the periodogram. The estimated mode parameters (frequency, amplitude and width) are comparable for the three power spectrum estimates, except for modes with very small mode widths (a few frequency bins), where the multitaper spectra broadened the modest compared with the periodogram. We tested the influence of the number of tapers used and found that narrow modes at low n values are broadened to the extent that they can no longer be fit if the number of tapers is too large. For helioseismic time series of this length and temporal resolution, the optimal number of tapers is less than 10.

Komm, R. W.↗

The SDO/EVE Solar Irradiance Coronal Dimming Index Catalog. I. Methods and Algorithms

When a coronal mass ejection departs, it leaves behind a temporary void. That void is known as coronal dimming, and it contains information about the mass ejection that caused it. Other physical processes can cause parts of the corona to have transient dimmings, but mass ejections are particularly interesting because of their influence in space weather. Prior work has established that dimmings are detectable even in disk-integrated irradiance observations, i.e., Sun-as-a-star measurements. The present work evaluates four years of continuous Solar Dynamics Observatory Extreme Ultraviolet Experiment (EVE) observations to greatly expand the number of dimmings we may detect and characterize, and collects that information into James’s EVE Dimming Index catalog. This paper details the algorithms used to produce the catalog, provides statistics on it, and compares it with prior work. The catalog contains 5051 potential events (rows), which correspond to all robustly detected solar eruptive events in this time period as defined by >C1 flares. Each row has a corresponding 27,349 elements of metadata and parameterizations (columns). In total, this catalog is the result of analyzing 7.6 million solar ultraviolet light curves.

James’s EVE Dimming Index (JEDI)↗

SAFEGUARDS ENVELOPE: PREVIOUS WORK AND EXAMPLES

The future expansion of nuclear power will require not just electricity production but fuel cycle facilities such as fuel fabrication and reprocessing plants. As large reprocessing facilities are built in various states, they must be built and operated in a manner to minimize the risk of nuclear proliferation. Process monitoring has returned to the spotlight as an added measure that can increase confidence in the safeguards of special nuclear material (SNM). Process monitoring can be demonstrated to lengthen the allowable inventory period by reducing accountancy requirements, and to reduce the false positive indications. The next logical step is the creation of a Safeguards Envelope, a set of operational parameters and models to maximize anomaly detection and inventory period by process monitoring while minimizing operator impact and false positive rates. A brief example of a rudimentary Safeguards Envelope is presented, and shown to detect synthetic diversions overlaying a measured processing plant data set. This demonstration Safeguards Envelope is shown to increase the confidence that no SNM has been diverted with minimal operator impact, even though it is based on an information sparse environment. While the foundation on which a full Safeguards Envelope can be built has been presented in historical demonstrations of process monitoring, several requirements remain yet unfulfilled. Future work will require reprocessing plant transient models, inclusion of “non-traditional” operating data, and exploration of new methods of identifying subtle events in transient processes.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

On the Statistical Mechanics of Mass Accommodation at Liquid–Vapor Interfaces

Here we propose a framework for describing the dynamics associated with the adsorption of small molecules to liquid-vapor interfaces using an intermediate resolution between traditional continuum theories that are bereft of molecular detail and molecular dynamics simulations that are replete with them. In particular, we develop an effective single particle equation of motion capable of describing the physical processes that determine thermal and mass accommodation probabilities. The effective equation is parametrized with quantities that vary through space away from the liquid-vapor interface. Of particular importance in describing the early time dynamics is the spatially dependent friction, for which we propose a numerical scheme to evaluate from molecular simulation. Taken together with potentials of mean force computable with importance sampling methods, we illustrate how to compute the mass accommodation coefficient and residence time distribution. Throughout, we highlight the case of ozone adsorption in aqueous solutions and its dependence on electrolyte composition.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

The Dissolution of Fe in HCl from the Ilmenite Concentrate; Evaluating the Effect of Operating Parameters and Mutual Interactions

Selective HCl dissolution of ilmenite components for obtaining Ti or titanium dioxide (TiO{sub 2}) has been highly recognized due to its advantages, greater environmental friendliness, and simplicity, compared to H{sub 2}SO{sub 4} and Cl{sub 2} methods. The effect of numerous parameters has been studied with the one-factor-at-a-time method. The present study aimed to evaluate the effect of key operation parameters, such as acid-to-solid ratio (A/S: 5 to 20 mL/g), reaction temperature (T: 70 °C to 100 °C), and acid concentration (A pct: 15 to 30 wt pct), on the dissolution of Fe in HCl solution with the minimum Ti losses to the leachate from its abundant, domestic, and low-cost mineral source (Kahnooj ilmenite concentrate) using central composite design–response surface methodology. After 90 minutes of leaching, the Ti/Fe (pct) in terms of dissolved amounts was selected as the process assessment response function. Based on the conducted experimental and statistical analysis, increasing the levels of parameters in the studied domain leads to an increase in Ti/Fe (pct), in the order of A pct > T > A/S. Two statistically significant mutual interactions between A/S-T and T-A pct, with 95 pct confidence level, were revealed for the first time in this study. The optimization strategy was set to the minimization of Ti/Fe (pct) by considering the objective of study and the selected response function. The A/S, T, and A pct were determined to be 5 mL/g, 70 °C, and 15 pct, respectively, for maximum impurity dissolution and minimum Ti loss to the leachate.

36 MATERIALS SCIENCE↗

An angular biasing method using arbitrary convex polyhedra for Monte Carlo radiation transport calculations

This paper presents a new method for performing angular biasing in Monte Carlo radiation transport codes using arbitrary convex polyhedra to define regions of interest toward which to project particles (DXTRAN regions). The method is derived and is implemented using axis-aligned right parallelepipeds (AARPPs) and arbitrary convex polyhedra. Attention is also paid to possible numerical complications and areas for future refinement. A series of test problems are executed with void, purely absorbing, purely scattering, and 50% absorbing/50% scattering materials. For all test problems tally results using AARPP and polyhedral DXTRAN regions agree with analog and/or spherical DXTRAN results within statistical uncertainties. In cases with significant scattering the figure of merit (FOM) using AARPP or polyhedral DXTRAN regions is lower than with spherical regions despite the ability to closely fit the tally region. This is because spherical DXTRAN processing is computationally less expensive than AARPP or polyhedral DXTRAN processing. Thus, it is recommended that the speed of spherical regions be considered versus the ability to closely fit the tally region with an AARPP or arbitrary polyhedral region. It is also recommended that short calculations be made prior to final calculations to compare the FOM for the various DXTRAN geometries because of the influence of the scattering behavior.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Model of delocalized atoms in the physics of the vitreous state

A development of the model of delocalized atoms of liquids and glasses is proposed. It is shown that the basic equation of the model for the probability of delocalization (excitation) of an atom can be obtained not only from the Clausius relation but also by other methods of statistical physics. Techniques for calculating the parameters of the model are developed. The critical displacement of an atom from the equilibrium position, which corresponds to the maximum interatomic attraction force, can be considered as a delocalization (local excitation) of this atom in an elastic continuum. The energy of the critical displacement of an atom calculated as the work of the limit elastic deformation of the interatomic bond in an elastic continuum is in agreement with the results of calculation by the model of delocalized atoms. This energy can also be calculated from the data on surface tension and atomic volume. In silicate glasses, the process of delocalization of an atom represents the critical displacement of a bridging oxygen atom in the structural fragment of a silicon-oxygen (Si-O-Si) network before the switching of the valence bond, whereas, in amorphous organic polymers, the delocalization of an atom corresponds to the limit displacement of a fragment of the main chain of a macromolecule (a group of atoms in the connecting link).

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Multifidelity Ensemble Kalman Filtering Using Surrogate Models Defined by Theory-Guided Autoencoders

Data assimilation is a Bayesian inference process that obtains an enhanced understanding of a physical system of interest by fusing information from an inexact physics-based model, and from noisy sparse observations of reality. The multifidelity ensemble Kalman filter (MFEnKF) recently developed by the authors combines a full-order physical model and a hierarchy of reduced order surrogate models in order to increase the computational efficiency of data assimilation. The standard MFEnKF uses linear couplings between models, and is statistically optimal in case of Gaussian probability densities. This work extends the MFEnKF into to make use of a broader class of surrogate model such as those based on machine learning methods such as autoencoders non-linear couplings in between the model hierarchies. We identify the right-invertibility property for autoencoders as being a key predictor of success in the forecasting power of autoencoder-based reduced order models. We propose a methodology that allows us to construct reduced order surrogate models that are more accurate than the ones obtained via conventional linear methods. Numerical experiments with the canonical Lorenz'96 model illustrate that nonlinear surrogates perform better than linear projection-based ones in the context of multifidelity ensemble Kalman filtering. We additionality show a large-scale proof-of-concept result with the quasi-geostrophic equations, showing the competitiveness of the method with a traditional reduced order model-based MFEnKF.

97 MATHEMATICS AND COMPUTING↗

SPITZER SECONDARY ECLIPSE DEPTHS WITH MULTIPLE INTRAPIXEL SENSITIVITY CORRECTION METHODS OBSERVATIONS OF WASP-13b, WASP-15b, WASP-16b, WASP-62b, AND HAT-P-22b

We measure the 4.5 μ m thermal emission of five transiting hot Jupiters, WASP-13b, WASP-15b, WASP-16b, WASP-62b, and HAT-P-22b using channel 2 of the Infrared Array Camera (IRAC) on the Spitzer Space Telescope . Significant intrapixel sensitivity variations in Spitzer IRAC data require careful correction in order to achieve precision on the order of several hundred parts per million (ppm) for the measurement of exoplanet secondary eclipses. We determine eclipse depths by first correcting the raw data using three independent data reduction methods. The Pixel Gain Map (PMAP), Nearest Neighbors (NNBR), and Pixel Level Decorrelation (PLD) each correct for the intrapixel sensitivity effect in Spitzer photometric time-series observations. The results from each methodology are compared against each other to establish if they reach a statistically equivalent result in every case and to evaluate their ability to minimize uncertainty in the measurement. We find that all three methods produce reliable results. For every planet examined here NNBR and PLD produce results that are in statistical agreement. However, the PMAP method appears to produce results in slight disagreement in cases where the stellar centroid is not kept consistently on the most well characterized area of the detector. We evaluate the ability of each method to reduce the scatter in the residuals as well as in the correlated noise in the corrected data. The NNBR and PLD methods consistently minimize both white and red noise levels and should be considered reliable and consistent. The planets in this study span equilibrium temperatures from 1100 to 2000 K and have brightness temperatures that require either high albedo or efficient recirculation. However, it is possible that other processes such as clouds or disequilibrium chemistry may also be responsible for producing these brightness temperatures.

79 ASTRONOMY AND ASTROPHYSICS↗

ATOMIC PARTITION FUNCTION FOR STELLAR ATMOSPHERES AND PLASMA DIAGNOSTICS

A new approximate partition function is derived as a function of temperature and total number density of particles in the given system, and three adjustable parameters. The derivation assumes that we can simulate the calculations of the partition function for hydrogen by means of averages of the energies and sums of the statistical weights. We present the procedure and mathematical process to obtain an approximate analytic function and its derivatives that depend on those parameters. The comparisons with other calculations reported in the literature show good agreement. The free parameters of this function are calculated and given in a table for all the ions of the first 20 atomic species.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A DETECTION OF H{alpha} IN AN EXOPLANETARY EXOSPHERE

We report on a search for H{alpha} absorption in four exoplanets. Strong features at H{alpha} are observed in the transmission spectra of both HD 189733b and HD 209458b. We attempt to characterize and remove the effects of stellar variability in HD 189733b, and along with an empirical Monte Carlo test the results imply a statistically significant transit-dependent feature of (- 8.72 {+-} 1.48) Multiplication-Sign 10{sup -4} integrated over a 16 A band relative to the adjacent continuum. We interpret this as the first detection of this line in an exoplanetary atmosphere. A previous detection of Ly{alpha} in HD 189733b's atmosphere allows us to calculate an excitation temperature for hydrogen, T{sub exc} = 2.6 Multiplication-Sign 10{sup 4} K. This calculation depends significantly on certain simplifying assumptions. We explore these assumptions and argue that T{sub exc} is very likely much greater than the radiative equilibrium temperature (the temperature the planet is assumed to be at based on stellar radiation and the planetary distance) of HD 189733b. A large T{sub exc} implies a very low density that is not in thermodynamic equilibrium with the planet's lower atmosphere. We argue that the n = 2 hydrogen required to cause H{alpha} absorption in the atmosphere is created as a result of the greater UV flux at HD 189733b, which has the smallest orbit and most chromospherically active central star in our sample. Though the overall integration of HD 209458b's transmission spectrum over a wide band is consistent with zero, it contains a dramatic, statistically significant feature in the transmission spectrum with reflectional symmetry. We discuss possible physical processes that could cause this feature. Our remaining two targets (HD 147506b and HD 149026b) do not show any clear features, so we place upper limits on their H{alpha} absorption levels.

79 ASTRONOMY AND ASTROPHYSICS↗

Empirical radius formulas for canonical neutron stars from bidirectionally selecting features of equations of state in extended Bayesian analyses of observational data

Significant advancement in Bayesian inference of nuclear equation of state (EOS) from gravitational wave and x-ray observations of neutron stars (NSs) has been made by the nuclear astrophysics community especially since GW170817. By extending the traditional Bayesian analysis which normally ends at presenting the marginalized posterior probability distribution functions (PDFs) of individual EOS parameters and their correlations (or sometimes only the Pearson correlation coefficients which are only reliably useful when the variables are linearly correlated while they are actually often not), we search for a data-driven and robust empirical formula for the radius 𝑅 1.4 of canonical NSs in terms of the characteristic EOS parameters (features). We also identify the single most important but currently poorly known EOS parameter for determining the 𝑅 1.4 . Using three regression-model-building methodologies: bidirectional stepwise feature selection, least absolute shrinkage selection operator (LASSO) regression, and neural network regression on a large set of posterior EOSs and the corresponding 𝑅 1.4 values inferred from earlier comprehensive Bayesian analyses of NS observational data, we systematically and rigorously develop the most probable 𝑅 1.4 formulas with varying statistical accuracy and technical complexity. Here, the most important EOS parameters for determining 𝑅 1.4 are found consistently in each of the feature selection processes to be (in order of decreasing importance): curvature 𝐾 sym , slope 𝐿, skewness 𝐽 sym of nuclear symmetry energy, skewness 𝐽 0 , incompressibility 𝐾 0 of symmetric nuclear matter, and the magnitude 𝐸 sym ⁡(𝜌 0 ) of symmetry energy at the saturation density 𝜌 0 of nuclear matter.

Bayesian methods↗

Automated fibroglandular tissue segmentation and volumetric density estimation in breast MRI using an atlas-aided fuzzy C-means method

Purpose: Breast magnetic resonance imaging (MRI) plays an important role in the clinical management of breast cancer. Studies suggest that the relative amount of fibroglandular (i.e., dense) tissue in the breast as quantified in MR images can be predictive of the risk for developing breast cancer, especially for high-risk women. Automated segmentation of the fibroglandular tissue and volumetric density estimation in breast MRI could therefore be useful for breast cancer risk assessment. Methods: In this work the authors develop and validate a fully automated segmentation algorithm, namely, an atlas-aided fuzzy C-means (FCM-Atlas) method, to estimate the volumetric amount of fibroglandular tissue in breast MRI. The FCM-Atlas is a 2D segmentation method working on a slice-by-slice basis. FCM clustering is first applied to the intensity space of each 2D MR slice to produce an initial voxelwise likelihood map of fibroglandular tissue. Then a prior learned fibroglandular tissue likelihood atlas is incorporated to refine the initial FCM likelihood map to achieve enhanced segmentation, from which the absolute volume of the fibroglandular tissue (|FGT|) and the relative amount (i.e., percentage) of the |FGT| relative to the whole breast volume (FGT%) are computed. The authors' method is evaluated by a representative dataset of 60 3D bilateral breast MRI scans (120 breasts) that span the full breast density range of the American College of Radiology Breast Imaging Reporting and Data System. The automated segmentation is compared to manual segmentation obtained by two experienced breast imaging radiologists. Segmentation performance is assessed by linear regression, Pearson's correlation coefficients, Student's pairedt-test, and Dice's similarity coefficients (DSC). Results: The inter-reader correlation is 0.97 for FGT% and 0.95 for |FGT|. When compared to the average of the two readers’ manual segmentation, the proposed FCM-Atlas method achieves a correlation ofr = 0.92 for FGT% and r = 0.93 for |FGT|, and the automated segmentation is not statistically significantly different (p = 0.46 for FGT% and p = 0.55 for |FGT|). The bilateral correlation between left breasts and right breasts for the FGT% is 0.94, 0.92, and 0.95 for reader 1, reader 2, and the FCM-Atlas, respectively; likewise, for the |FGT|, it is 0.92, 0.92, and 0.93, respectively. For the spatial segmentation agreement, the automated algorithm achieves a DSC of 0.69 ± 0.1 when compared to reader 1 and 0.61 ± 0.1 for reader 2, respectively, while the DSC between the two readers’ manual segmentation is 0.67 ± 0.15. Additional robustness analysis shows that the segmentation performance of the authors' method is stable both with respect to selecting different cases and to varying the number of cases needed to construct the prior probability atlas. The authors' results also show that the proposed FCM-Atlas method outperforms the commonly used two-cluster FCM-alone method. The authors' method runs at ∼5 min for each 3D bilateral MR scan (56 slices) for computing the FGT% and |FGT|, compared to ∼55 min needed for manual segmentation for the same purpose. Conclusions: The authors' method achieves robust segmentation and can serve as an efficient tool for processing large clinical datasets for quantifying the fibroglandular tissue content in breast MRI. It holds a great potential to support clinical applications in the future including breast cancer risk assessment.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Physics-informed semantic inpainting: Application to geostatistical modeling

A fundamental problem in geostatistical modeling is to infer the heterogeneous geological field based on limited measurements and some prior spatial statistics. Semantic inpainting, a technique for image processing using deep generative models, has been recently applied for this purpose, demonstrating its effectiveness in dealing with complex spatial patterns. However, the original semantic inpainting framework incorporates only information from direct measurements, while in geostatistics indirect measurements are often plentiful. In this work, to overcome this limitation, we propose a physics-informed semantic inpainting framework, employing the Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) and jointly incorporating the direct and indirect measurements by exploiting the underlying physical laws. Our simulation results for a high-dimensional problem with 512 dimensions show that in the new method, the physical conservation laws are satisfied and contribute in enhancing the inpainting performance compared to using only the direct measurements.

54 ENVIRONMENTAL SCIENCES↗

Understanding GPU Power. A Survey of Profiling, Modeling, and Simulation Methods

Modern graphics processing units (GPUs) have complex architectures that admit exceptional performance and energy efficiency for high throughput applications.Though GPUs consume large amounts of power, their use for high throughput applications facilitate state-of-the-art energy efficiency and performance. Consequently, continued development relies on understanding their power consumption. Our work is a survey of GPU power modeling and profiling methods with increased detail on noteworthy efforts. Moreover, as direct measurement of GPU power is necessary for model evaluation and parameter initiation, internal and external power sensors are discussed. Hardware counters, which are low-level tallies of hardware events, share strong correlation to power use and performance. Statistical correlation between power and performance counters has yielded worthwhile GPU power models, yet the complexity inherent to GPU architectures presents new hurdles for power modeling. Developments and challenges of counter-based GPU power modeling is discussed. Often building on the counter-based models, research efforts for GPU power simulation, which make power predictions from input code and hardware knowledge, provide opportunities for optimization in programming or architectural design. Noteworthy strides in power simulations for GPUs are included along with their performance or functional simulator counterparts when appropriate. Lastly, possible directions for future research are discussed.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗