Publications

Characterizing shielding effectiveness (SE) of enclosures is important in aerospace, military, and consumer applications. Direct SE measurement of an enclosure or chassis may be considered an exact characterization, but there are several sources of possible variability in such measurements, e.g., mechanical tolerances, the absence of components during test that exist in a final assembly, movement of components and cables, and perturbations due to probes and associated cabling. In [1] , internal stirrers were investigated as a way to sample the variation of SE of small enclosures when populated with random metallic objects. Here, we explore this idea as a way to quantify variability and sensitivity of an SE measurement, not only indicating the uncertainty of the SE measurement, but also delineating frequency ranges where either deterministic or statistical simulations should be applied.

There has always been a desire to port high-fidelity reactive flow models from one code to another. For example, the AWE reactive burn model known as CREST has been or is being implemented in several of the U.S. Department of Energy hydrocodes. Those involved with reactive burn model implementation recognize the challenges immediately, e.g., Eulerian versus Lagrangian frameworks, the form of the equation of state, the closure relations, etc. In this work, we report the development of the CREST reactive burn model in CTH, a multidimensional, multi-material hydrocode developed by Sandia National Laboratories, following an earlier implementation shown at the last International Detonation Symposium. Results include code-to-code comparisons between CTH and the AWE hydrocode PERUSE, focusing on the simulated particle velocity histories during a shock-to-detonation transition, and corresponding to previous gas gun impact experiments as well as new model verification studies. Lessons learned are provided, including discussions of the numerical accuracy, in addition to the role of artificial viscosity and artificial viscous work. Finally, simulation results are shown to compare the Snowplough versus P-Alpha porosity model options.

This paper presents the ongoing development of a chloride-based wireline tool designed to detect and quantify inflows from feed zones in geothermal wells. The tool aims to characterize stimulation events in EGS wells at Utah FORGE (Frontier Observatory for Research in Geothermal Energy) and other EGS sites. Successful development of the chloride tool would greatly improve production monitoring of the fractures and enable proactive prescription of additional stimulations over the life of the field, thus helping to improve EGS commercial feasibility. The recent developments of the chloride tool have focused on preparing for and conducting the field deployment at the Utah FORGE site. The field-scale tool assembly features a FORGE sensor package housing the Ion Selective Electrode (ISE), a pH electrode, and a reference electrode, as well as a Mitco PTS sensor package for secondary downhole measurements. A high-temperature logging tool has been developed and tested to capture and transmit data from the chemical sensors to the surface through a 7-conductor wireline cable. Alongside the development of the field-scale tool, flow experiments were carried out in the artificial well system at the Stanford Geothermal Lab. These experiments provided crucial insights into how the chemical tool responds to different variables, including the chloride concentration in the feed zone, its vertical positioning relative to the feed zone, and the presence of other chemical species in the feed zone fluid. The results highlight the tool's sensitivity to various parameters, underscoring the potential of using chloride concentration measurements as a method for inferring feed zone inflow rates in geothermal wells. The tool was successfully deployed at the Utah FORGE site using a wireline truck in the vertical well 58-32 and the directional production well 16B(78)-32.

A new Adaptive Mesh Refinement (AMR) keyword was added to the CTH¹ hydrocode developed at Sandia National Laboratories (SNL). The new indicator keyword, "ratec*ycle", allows the user to specify the minimum number of computational cycles before an AMR block is allowed to be un-refined. This option is designed to allow the analyst to control how quickly a block is un-refined to avoid introducing anomalous waves in their solution due to information propagating across mesh resolution changes. For example, in reactive flow simulations it is often desirable to accurately capture the expansion region behind the reaction front. The effect of this new option was examined using the XHVRB^{2, 3} model for XTX8003 to model the propagation of the detonation wave in explosives in small channels, and also for a simpler explosive model driving a steel case. The effect on computational cost as a function of this new option was also examined.

Multiple scattering is a common phenomenon in acoustic media that arises from the interaction of the acoustic field with a network of scatterers. This mechanism is dominant in problems such as the design and simulation of acoustic metamaterial structures often used to achieve acoustic control for sound isolation, and remote sensing. In this study, we present a physics-informed neural network (PINN) capable of simulating the propagation of acoustic waves in an infinite domain in the presence of multiple rigid scatterers. This approach integrates a deep neural network architecture with the mathematical description of the physical problem in order to obtain predictions of the acoustic field that are consistent with both governing equations and boundary conditions. The predictions from the PINN are compared with those from a commercial finite element software model in order to assess the performance of the method.

Tabulated chemistry models are widely used to simulate large-scale turbulent fires in applications including energy generation and fire safety. Tabulation via piecewise Cartesian interpolation suffers from the curse-of-dimensionality, leading to a prohibitive exponential growth in parameters and memory usage as more dimensions are considered. Artificial neural networks (ANNs) have attracted attention for constructing surrogates for chemistry models due to their ability to perform high-dimensional approximation. However, due to well-known pathologies regarding the realization of suboptimal local minima during training, in practice they do not converge and provide unreliable accuracy. Partition of unity networks (POUnets) are a recently introduced family of ANNs which preserve notions of convergence while performing high-dimensional approximation, discovering a mesh-free partition of space which may be used to perform optimal polynomial approximation. We assess their performance with respect to accuracy and model complexity in reconstructing unstructured flamelet data representative of nonadiabatic pool fire models. Our results show that POUnets can provide the desirable accuracy of classical spline-based interpolants with the low memory footprint of traditional ANNs while converging faster to significantly lower errors than ANNs. For example, we observe POUnets obtaining target accuracies in two dimensions with 40 to 50 times less memory and roughly double the compression in three dimensions. We also address the practical matter of efficiently training accurate POUnets by studying convergence over key hyperparameters, the impact of partition/basis formulation, and the sensitivity to initialization.

I experimentally investigated and modeled a proposed frequency-domain method for detecting and tracking cislunar spacecraft and near-earth asteroids using heliostat fields at night. Unlike imaging, which detects spacecraft and asteroids by their streak in star-fixed long-exposure photographs, the proposed detection method oscillates the orientation of heliostats concentrating light from the stellar field and measures the light's photocurrent power spectrum at sub-milliHertz resolution. If heliostat oscillation repetitively traces out a closed loop fixed to the stars, light from spacecraft or asteroids moving along that loop produce photocurrent at a frequency slightly shifted from starlight. The frequency shift is proportional to the spacecraft or asteroid's apparent angular rate relative to sidereal. Relative phase corresponds to relative angular position, enabling tracking. Since heliostats are inexpensive compared to an astronomical observatory and otherwise unused at night, the proposed method may cost-effectively augment observatory systems such as NASA's Asteroid Terrestrial-impact Last Alert System (ATLAS).

Different data pipelines and statistical methods are applied to photovoltaic (PV) performance datasets to quantify the performance loss rate (PLR). Since the real values of PLR are unknown, a variety of unvalidated values are reported. As such, the PV industry commonly assumes PLR based on statistically extracted ranges from the literature. However, the accuracy and uncertainty of PLR depend on several parameters including seasonality, local climatic conditions, and the response of a particular PV technology. In addition, the specific data pipeline and statistical method used affect the accuracy and uncertainty. To provide insights, a framework of (≈200 million) synthetic simulations of PV performance datasets using data from different climates is developed. Time series with known PLR and data quality are synthesized, and large parametric studies are conducted to examine the accuracy and uncertainty of different statistical approaches over the contiguous US, with an emphasis on the publicly available and “standardized” library, RdTools. In the results, it is confirmed that PLRs from RdTools are unbiased on average, but the accuracy and uncertainty of individual PLR estimates vary with climate zone, data quality, PV technology, and choice of analysis workflow. Best practices and improvement recommendations based on the findings of this study are provided.

Gaseous hydrogen is known to embrittle most steels, including the steels used in natural gas pipelines. As injection of hydrogen into the existing natural gas infrastructure is considered globally by the pipeline industry, the structural integrity of pipelines transporting gaseous hydrogen must be investigated. Hydrogen Extremely Low Probability of Rupture (HELPR) is a publicly available and open-source probabilistic fatigue and fracture mechanics toolkit recently developed at Sandia National Laboratories. HELPR is intended to incorporate the influence of hydrogen into structural integrity assessments of natural gas transmission and distribution infrastructure. HELPR utilizes engineering models, such as those specified in ASME B31.12 and API 579, with relatively low computational costs to perform large sample ensembles, enabling estimation of performance distributions including low probability tail estimates. Leveraging the probabilistic capabilities built into HELPR, the sensitivity of fatigue and fracture calculations to specific modeling parameters on performance margins can be quantified. Through applying HELPR’s probabilistic capabilities to realistic scenarios, the impact of uncertainty in specific model parameter descriptions on performance margins, such as cycles to unstable crack growth or rupture in gaseous hydrogen, can be characterized; this same approach can then be used to assess the impact of reducing uncertainty sources on the resulting performance metrics, margins, and associated risks. A few industry-motivated scenarios are used to demonstrate this approach.

Evaluation of the hydrogen storage capacity of porous rocks is crucial for underground hydrogen storage. Using 1H nuclear magnetic resonance (NMR) spectroscopy, we successfully characterized the hydrogen responses and identified storage mechanisms in Berea sandstone under varying water saturation. The results indicate that the injected hydrogen behaves as a free gas phase and is capable of occupying the empty pore volume regardless of the saturation state. No hysteresis was observed during injection and production cycles.

With the advent of grid-forming inverters, the question of how they behave during and immediately following short-circuit faults has become of major importance in the choice of control strategy and inverter design. This problem has greatly stimulated activity in the study of power electronics and their effects on large-scale systems, as evidenced by the formation of the Universal Interoperability for Grid-Forming Inverters (UNIFI) Consortium. This paper investigates a particular droop-controlled grid-forming inverter model called REGFM-Al, recently ap-proved for use in system-wide studies by the WECC Modeling and Validation Subcommittee. We show that following 3-phase faults, inverters that behave according to the REGFM-Al specification may inject significant post-fault active power transients. We describe the mechanism by which this occurs and propose a state reset strategy for mitigating the amplitude of these transients. To illustrate the main concepts, we present simulation results for a two-area test system augmented with a 200MVA solar photovoltaic power plant represented as a single equivalent grid-forming inverter. The analysis described in this paper was performed in MATLAB using the Power and Energy Storage Systems Toolbox (PSTess). To validate the model implementation in PSTess, we compared simulation results against GE Vernova's Positive Sequence Load Flow (PSLF) software.

Over the past few years, advancements in closed-loop geothermal systems (CLGS), also called advanced geothermal systems (AGS), have sparked a renewed interest in these types of designs. CLGS have certain advantages over traditional and enhanced geothermal systems (EGS), including not requiring in-situ reservoir permeability, conservation of the circulating fluid, and allowing for different fluids, including working fluids directly driving a turbine at the surface. CLGS may be attractive in environments where water resources are limited, rock contaminants must be avoided, and stimulation treatments are not available (e.g., due to regulatory or technical reasons). Despite these advantages, CLGS have some challenges, including limited surface area for heat transfer and requiring long wellbores and laterals to obtain multi-MW output in conduction-only reservoirs. CLGS have been investigated in conduction-only systems. In this paper, we explore the impact of both forced and natural convection on the levels of heat extraction with a CLGS deployed in a hot wet rock reservoir. We bound potential benefits of convection by investigating liquid reservoirs over a range of natural and forced convective coefficients. Additionally, we investigate the effects of permeability, porosity, and geothermal temperature gradient in the reservoir on CLGS outputs. Reservoir simulations indicate that reservoir permeabilities of at least ~100 mD are required for natural convection to increase the heat output with respect to a conduction-only scenario. The impact increases with increasing reservoir temperature. When subject to a forced convection flow field, Darcy velocities of at least 10-7 m/s are required to obtain an increase in heat output.

The eXtended History Variable Reactive Burn (XHVRB) model is parameterized for hexanitrostilbene (HNS) and pentaerythritol tetranitrate (PETN) based on data collected from a series of high-throughput initiation (HTI) experiments. The HTI experiment has generated a wealth of thin-pulse, sub-millimeter shock initiation data for a variety of vapor deposited explosive films. This is because it provides access to growth-to-detonation information for explosives that exhibit a shock-to-detonation transition (SDT) with length and time scales that are too short to be resolved by conventional experiments. The XHVRB model was selected because previous work has shown that simpler, homogeneous reactive burn models (RBMs) were incapable of reproducing the particle velocity buildup observed in experiments with heterogeneous explosives. Therefore, calibrated XHVRB models are developed in an attempt to capture the heterogeneous behavior not captured previously and to assess the models’ ability to capture the SDT across multiple explosive film thicknesses.

Intermolecular Coulombic decay (ICD) in liquid water is a relatively novel type of nonlocal electronic decay mechanism, competing with the traditional mechanism of proton transfer between neighboring water molecules. Key features of ICD are its ultrafast non-radiative decay process and ultralong-range for excess energy transfer from the excited atom/molecule to its neighbors. Since detecting unambiguous ICD signatures in bulk liquid water is technically challenging, small water clusters have often been utilized to gain insights into ICD and other ionization processes in aqueous environment. Here, we present results from quantum mechanical calculations of the electronic structures of neutral to multiply-ionized water monomer, dimer, trimer, and tetramer. Core-level electrons of water are also considered here since recent studies demonstrated that emission site and energy of the electrons released during resonant-Auger-ICD cascade can be controlled by coupling ICD to resonant core excitation. Previous studies of ICD and electronic structures of neutral and ionized small water clusters and liquid water are briefly discussed.

Closed-loop geothermal systems (CLGSs) rely on circulation of a heat transfer fluid in a closed-loop design without penetrating the reservoir to extract subsurface heat and bring it to the surface. We developed and applied numerical models to study u-shaped and coaxial CLGSs in hot-dry-rock over a more comprehensive parameter space than has been studied before, including water and supercritical CO2 (sCO2) as working fluids. An economic analysis of each realization was performed to evaluate the levelized cost of heat (LCOH) for direct heating application and levelized cost of electricity (LCOE) for electrical power generation. The results of the parameter study, composed of 2.5 million simulations, combined with a plant and economic model comprise the backbone of a publicly accessible web application that can be used to query, analyze, and plot outlet states, thermal and mechanical power output, and LCOH/LCOE, thereby facilitating feasibility studies led by potential developers, geothermal scientists, or the general public (https://gdr.openei.org/submissions/1473). Our results indicate competitive LCOH can be achieved; however, competitive LCOE cannot be achieved without significant reductions in drilling costs. We also present a site-based case study for multi-lateral systems and discuss how our comprehensive single-lateral analyses can be applied to approximate multi-lateral CLGSs. Looking beyond hot-dry-rock, we detail CLGS studies in permeable wet rock, albeit for a more limited parameter space, indicating that reservoir permeability of greater than 250 mD is necessary to significantly improve CLGS power production, and that reservoir temperatures greater than 200 °C, achieved by going to greater depths (∼3–4 km), may significantly enhance power production.

A mesoscale model for the shock initiation of pentaerythritol tetranitrate (PETN) films has been utilized to elucidate changes in initiation thresholds due to aging conditions and surface roughness, as has been observed from a series of high-throughput initiation (HTI) experiments. The HTI experiment has generated a wealth of thin-pulse, sub-millimeter shock initiation data for vapor deposited PETN films with thicknesses of 67-125 μm and varying accelerated aging conditions. This is because the HTI experiment provides access to growth-to-detonation information for explosives that exhibit a shock-to-detonation transition (SDT) with length and time scales that are too short to be resolved by conventional experiments. Mesoscale modeling results using experimentally characterized PETN microstructures are able to capture the general trend observed in experiments, in that increasing flyer impact velocity increases reactions until full detonation is reached. Moreover, the varying degrees of surface roughness that were considered were found to provide only minor variances in the peak particle velocity at the explosive output. The model did not predict a shift in the initiation threshold due to aged microstructures alone, indicating that additional mesoscale model improvements are necessary.

In low inertia grids, significant frequency deviations can occur as a result of changes in power (load, generation, etc.), These deviations may activate various protection schemes designed to safeguard the system, potentially leading to blackouts. Therefore, assessing the frequency stability of the power system is crucial. The Frequency Security Index (FSI) serves as a metric for evaluating system stability. However, computing the FSI for a specific load change necessitates actual load changes on the system, which is often impractical. This paper introduces a method for calculating the FSI without requiring load changes for all values. A mathematical expression for the FSI is derived, which uses the values of microgrid parameters (such as inertia and damping constant) to compute the FSI for any load change. Subsequently, the parameters that most significantly affect the FSI are identified. Then, the paper introduces a Moving Horizon Estimation (MHE)-based parameter estimation approach, which leverages small perturbations from an energy storage system to estimate the most influential parameters for the FSI. The results show that the FSI calculation with the estimated parameters is more accurate (compared to COI averaged parameters), enabling a more effective state of health monitoring of the microgrid.

The development of multi-axis force sensing ca-pabilities in elastomeric materials has enabled new types of human motion measurement with many potential applications. In this work, we present a new soft insole that enables mobile measurement of ground reaction forces (GRFs) outside of a lab-oratory setting. This insole is based on hybrid shear and normal force detecting (SAND) tactile elements (taxels) consisting of optical sensors optimized for shear sensing and piezoresistive pressure sensors dedicated to normal force measurement. We develop polynomial regression and deep neural network (DNN) GRF prediction models and compare their performance to ground-truth force plate data during two walking experiments. Utilizing a 4-layer DNN, we demonstrate accurate prediction of the anterior-posterior (AP), medial-lateral (ML) and vertical components of the GRF with normalized mean absolute errors (NMAE) of <5.1 %, 4.1 %, and 4.5%, respectively. We also demonstrate the durability of the hybrid SAND insole construction through more than 20,000 cycles of use.

A method for battery state of charge (SoC) estimation that compensates input noise using an adaptive square-root unscented Kalman filter (ASRUKF) is presented in this paper. In contrast to traditional state estimation approaches that consider deterministic system inputs, this method can improve the accuracy of battery state estimator by considering that the measurements of the control input variable of the filter, the cell currents, are subject to noise. Also, this paper presents two estimators for input and output noise covariance. The proposed method consists of initialization, state correction, sigma point calculations, state prediction, and covariance estimation steps and is demonstrated using simulations. We simulate two battery cycling protocols of three series-connected batteries whose SoC is estimated by the proposed method. The results show that the improved ASRUKF can track closely the states and achieves a 20.63 % reduction in SoC estimation error when compared to a benchmark that does not consider input noise.

This paper provides a summary of planning work for experiments that will be necessary to address the long-term model validation needs required to meet offshore wind energy deployment goals. Conceptual experiments are identified and laid out in a validation hierarchy for both wind turbine and wind plant applications. Instrumentation needs that will be required for the offshore validation experiments to be impactful are then listed. The document concludes with a nominal vision for how these experiments can be accomplished.

As deep learning networks increase in size and performance, so do associated computational costs, approaching prohibitive levels. Dendrites offer powerful nonlinear "on-The-wire"computational capabilities, increasing the expressivity of the point neuron while preserving many of the advantages of SNNs. We seek to demonstrate the potential of dendritic computations by combining them with the low-power event-driven computation of Spiking Neural Networks (SNNs) for deep learning applications. To this end, we have developed a library that adds dendritic computation to SNNs within the PyTorch framework, enabling complex deep learning networks that still retain the low power advantages of SNNs. Our library leverages a dendrite CMOS hardware model to inform the software model, which enables nonlinear computation integrated with snnTorch at scale. By leveraging dendrites in a deep learning framework, we examine the capabilities of dendrites via coincidence detection and comparison in a machine learning task with a SNN. Finally, we discuss potential deep learning applications in the context of current state-of-The-Art deep learning methods and energy-efficient neuromorphic hardware.

In many applications, one can only access the inexact gradients and inexact hessian times vector products. Thus it is essential to consider algorithms that can handle such inexact quantities with a guaranteed convergence to solution. An inexact adaptive and provably convergent semismooth Newton method is considered to solve constrained optimization problems. In particular, dynamic optimization problems, which are known to be highly expensive, are the focus. A memory efficient semismooth Newton algorithm is introduced for these problems. The source of efficiency and inexactness is the randomized matrix sketching. Applications to optimization problems constrained by partial differential equations are also considered.

Network interface controllers (NICs) with general-purpose compute capabilities ('SmartNICs') present an opportunity for reducing host application overheads by offloading non-critical tasks to the NIC. In addition to moving computation, offloading requires that associated data is also transferred to the NIC. To meet this need, we introduce a high-performance, general-purpose data movement service that facilitates the of-floading of tasks to SmartNICs: The SmartNIC Data Movement Service (SDMS). SDMS provides near-line-rate transfer band-widths between the host and NIC. Moreover, SDMS's In-transit Data Placement (IDP) feature can reduce (or even eliminate) the cost of serializing data on the NIC by performing the necessary data formatting during the transfer. To illustrate these capabilities, we provide an in-depth case study using SDMS to offload data management operations related to Apache Arrow, a popular data format standard. For single-column tables, SDMS can achieve more than 87% of baseline throughput for data buffers that are 128 KiB or larger (and more than 95% of baseline throughput for buffers that are 1 MiB or larger) while also nearly eliminating the host and SmartNIC overhead associated with Arrow operations.

Glass wedges are used increase the dimensionality of various optical measurements. Light refracted through the wedges can be focused to closely spaced points, lines or planes as shown in the applications herein.

Traditional electronics assemblies are typically packaged using physically or chemically blown potted foams to reduce the effects of shock and vibration. These potting materials have several drawbacks including manufacturing reliability, lack of internal preload control, and poor serviceability. A modular foam encapsulation approach combined with additively manufactured (AM) silicone lattice compression structures can address these issues for packaged electronics. These preloaded silicone lattice structures, known as foam replacement structures (FRSs), are an integral part of the encapsulation approach and must be properly characterized to model the assembly stresses and dynamics. In this study, dynamic test data is used to validate finite element models of an electronics assembly with modular encapsulation and a direct ink write (DIW) AM silicone FRS. A variety of DIW compression architectures are characterized, and their nominal stress-strain behavior is represented with hyperfoam constitutive model parameterizations. Modeling is conducted with Sierra finite element software, specifically with a handoff from assembly preloading and uniaxial compression in Sierra/Solid Mechanics to linear modal and vibration analysis in Sierra/Structural Dynamics. This work demonstrates the application of this advanced modeling workflow, and results show good agreement with test data for both static and dynamic quantities of interest, including preload, modal, and vibration response.