Machine learning-based data-driven modeling can allow computationally efficient time-dependent solutions of PDEs, such as those that describe subsurface multiphysical problems. In this work, our previous approach (Kadeethum et al., 2021d) of conditional generative adversarial networks (cGAN) developed for the solution of steady-state problems involving highly heterogeneous material properties is extended to time-dependent problems by adopting the concept of continuous cGAN (CcGAN). The CcGAN that can condition continuous variables is developed to incorporate the time domain through either element-wise addition or conditional batch normalization. Moreover, this framework can handle training data that contain different timestamps and then predict timestamps that do not exist in the training data. As a numerical example, the transient response of the coupled poroelastic process is studied in two different permeability fields: Zinn & Harvey transformation and a bimodal transformation. The proposed CcGAN uses heterogeneous permeability fields as input parameters while pressure and displacement fields over time are model output. Our results show that the model provides sufficient accuracy with computational speed-up. This robust framework will enable us to perform real-time reservoir management and robust uncertainty quantification in poroelastic problems.
Electrons in a semiconductor occupy states within certain energy ranges, called energy bands. The position of the Fermi level with respect to these energy bands determines the charge carrier type of the semiconductor. Molybdenum disulfide (MoS2) is a two-dimensional, n-type semiconductor with potential applications in flexible electronics, transparent electronics, and optoelectronics. Electronic devices containing MoS2 could be used in environments where radiation affects device performance. Thus, it is important to determine the impact of radiation on MoS2. A one-molecule-thick layer of MoS2 (monolayer) and a two-molecule-thick layer of MoS2 (bilayer) were placed onto different areas of a gold (Au) substrate containing 1.2-µm-deep holes. The MoS2 was suspended over these holes but supported by the Au elsewhere on the substrate. This sample configuration was used to determine the effect of He+ radiation on the electronic properties of the suspended MoS2 and the Au-supported MoS2. The MoS2 was irradiated by He+ ions in two stages. The energy bands of the MoS2 were measured with respect to the Fermi level via photoelectron emission microscopy before irradiation and after each irradiation stage. From each measurement, the charge carrier type of the MoS2 after the corresponding irradiation stage was determined. The Fermi levels of the suspended monolayer and bilayer decreased by ≈0.15 eV with respect to the bands during the first irradiation stage During the second irradiation stage, however, the Fermi levels didn’t change significantly. This lack of change supports the existence of a radiation threshold, above which the electronic properties of suspended MoS2 remain the same. The Fermi levels of the supported monolayer and bilayer increased over the cumulative irradiation and didn’t show evidence of a threshold. Thus, suspended MoS2 becomes less n-type as it is irradiated. Supported MoS2, however, becomes more n-type as it is irradiated. These results could inform the development of radiation tolerance standards for MoS2, and thus, radiation-tolerant MoS2-based electronics.
PyApprox is a Python-based one-stop-shop for probabilistic analysis of scientific numerical models. Easy to use and extendable tools are provided for constructing surrogates, sensitivity analysis, Bayesian inference, experimental design, and forward uncertainty quantification. The algorithms implemented represent the most popular methods for model analysis developed over the past two decades, including recent advances in multi-fidelity approaches that use multiple model discretizations and/or simplified physics to significantly reduce the computational cost of various types of analyses. Simple interfaces are provided for the most commonly-used algorithms to limit a user’s need to tune the various hyper-parameters of each algorithm. However, more advanced work flows that require customization of hyper-parameters is also supported. An extensive set of Benchmarks from the literature is also provided to facilitate the easy comparison of different algorithms for a wide range of model analyses. This paper introduces PyApprox and its various features, and presents results demonstrating the utility of PyApprox on a benchmark problem modeling the advection of a tracer in ground water.
Villa, Daniel V.; Schostek, Tyler; Bianchi, Carlo; Macmillan, Madeline; Carvallo, Juan P.
The Multi-scenario extreme weather simulator (MEWS) is a stochastic weather generation tool. The MEWS algorithm uses 50 or more years of National Oceanic and Atmospheric Association (NOAA) daily summaries [1] for maximum and minimum temperature and NOAA climate norms [2] to calculate historical heat wave and cold snap statistics. The algorithm takes these statistics and shifts them according to multiplication factors provided in the Intergovernmental Panel on Climate Change (IPCC) physical basis technical summary [3] for heat waves.
ERMA is leveraging Sandia’s Microgrid Design Toolkit (MDT) [1] and adding significant new features to it. Development of the MDT was primarily funded by the Department of Energy, Office of Electricity Microgrid Program with some significant support coming from the U.S. Marine Corps. The MDT is a software program that runs on a Microsoft Windows PC. It is an amalgamation of several other software capabilities developed at Sandia and subsequently specialized for the purpose of microgrid design. The software capabilities include the Technology Management Optimization (TMO) application for optimal trade-space exploration, the Microgrid Performance and Reliability Model (PRM) for simulation of microgrid operations, and the Microgrid Sizing Capability (MSC) for preliminary sizing studies of distributed energy resources in a microgrid.
The COVID-19 pandemic has forced many organizations—from national laboratories to private companies—to change their workforce model to incorporate remote work. This study and the summarized results sought to understand the experiences of remote workers and the ways that remote work can impact recruitment and retention, employee engagement, and career development. Sandia, like many companies, has committed to establishing a hybrid work model that will persist postpandemic, and more Sandia employees than ever before have initiated remote work agreements. This parallels the nationwide increase in remote employment and motivates this study on remote work as an enduring part of workforce models.