Search NASA⌕ Search

SEARCH · Search NASA

Results for “sequential experiments”

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 379 records · Page 21

Kinetics and Mechanism for Hydrothermal Conversion of Polyhydroxybutyrate (PHB) for Wastewater Valorization

Conventional wastewater treatment processes can be tailored to recover organic carbon from wastewater as intracellular polyhydroxybutyrate (PHB) polymer granules while simultaneously meeting effluent discharge standards. Traditional applications of PHB as a bioplastic are hampered by its suboptimal properties (e.g., brittle), lack of efficient and sustainable approaches for recovering PHB from cells, and concerns about wastewater-derived impurities. In this study, we report on the conversion of PHB and its monomer acids - 3-hydroxybutyric acid (3HBA) and crotonic acid (CA) - under hydrothermal conditions (in condensed water at elevated temperature and pressure) to form propylene, a valuable chemical intermediate that self-separates from water. PHB depolymerization results in a mixture of 3HBA and CA, which can interconvert via (de)hydration reactions that vary with prevailing reaction conditions. Further hydrothermal conversion of the monomer acids yields propylene and CO2. Conversion of 3HBA occurs at lower temperatures than CA, and a new concerted dehydration-decarboxylation pathway is proposed, which differs from the sequential dehydration (3HBA to CA) and decarboxylation (CA to propylene and CO2) pathway reported for dry thermal conversion. A kinetics network model informed by experimental results reveals that CA conversion to propylene and CO2 proceeds predominantly via hydration to 3HBA followed by the concerted dehydration-decarboxylation pathway rather than by direct decarboxylation of CA. Demonstrative experiments using PHB-containing methanotrophic biomass show results consistent with the model, producing propylene at near-theoretical yields at lower temperatures than reported previously.

09 BIOMASS FUELS↗

Constraint Optimization Problem For The Cutting Of A Cobalt Chrome Refractory Material

This paper shows a complete approach to solve a given problem, from the experimentation to the optimization of different cutting parameters. In response to an industrial problem of slotting FSX 414, a Cobalt-based refractory material, we have implemented a design of experiment to determine the most influent parameters on the tool life, the surface roughness and the cutting forces. After theses trials, an optimization approach has been implemented to find the lowest manufacturing cost while respecting the roughness constraints and cutting force limitation constraints. The optimization approach is based on the Response Surface Method (RSM) using the Sequential Quadratic programming algorithm (SQP) for a constrained problem. To avoid a local optimum and to obtain an accurate solution at low cost, an efficient strategy, which allows improving the RSM accuracy in the vicinity of the global optimum, is presented. With these models and these trials, we could apply and compare our optimization methods in order to get the lowest cost for the best quality, i.e. a satisfying surface roughness and limited cutting forces.

36 MATERIALS SCIENCE↗

Above-threshold multiphoton photoemission from noble metal surfaces

Exciting solids with intense femtosecond laser pulses prompts electrons of the interrogated material to respond in a highly nonlinear manner, as is evident in the emission of high-order harmonic radiation and photoelectrons with kinetic energies well above that of the driving photons. Such high-field interactions can be resolved, for example, in above-threshold multiphoton photoemission (ATP) spectroscopy. In this work, we interrogate the nonlinear photoelectric responses of the pristine copper, silver, and gold noble metal surfaces in (111) and (100) crystal orientations in the perturbative regime. Using multiphoton photoemission spectroscopy (mPP) excited by finely tuned optical fields, we characterize enhancement of the mPP and ATP yields from (111) surfaces in selected k||-momentum ranges when the occupied Shockley surface (SS) states are (near-)resonantly coupled by multiphoton transitions to image potential (IP) intermediate states in the excitation process. The ATP signal from the IP states of (111) surfaces is largely defined by their formation through polarization of SS electrons; this observation is contrasted with ATP experiments from the Ag(100) surface, for which the SS becomes an unoccupied resonance and the IP states can only be excited from bands with significantly more bulk character. In addition, based on the optical power and nonlinear order-dependent mPP spectra, we provide evidence for ATP being a one-step, rather than a sequential process, as previously postulated.

74 ATOMIC AND MOLECULAR PHYSICS↗

Near-equilibrium growth from borophene edges on silver

Two-dimensional boron, borophene, was realized in recent experiments but still lacks an adequate growth theory for guiding its controlled synthesis. Combining ab initio calculations and experimental characterization, we study edges and growth kinetics of borophene on Ag(111). In equilibrium, the borophene edges are distinctly reconstructed with exceptionally low energies, in contrast to those of other two-dimensional materials. Away from equilibrium, sequential docking of boron feeding species to the reconstructed edges tends to extend the given lattice out of numerous polymorphic structures. Furthermore, each edge can grow via multiple energy pathways of atomic row assembly due to variable boron-boron coordination. These pathways reveal different degrees of anisotropic growth kinetics, shaping borophene into diverse elongated hexagonal islands in agreement with experimental observations in terms of morphology as well as edge orientation and periodicity. These results further suggest that ultrathin borophene ribbons can be grown at low temperature and low boron chemical potential.

Science & Technology - Other Topics↗

Connected and Learning Based Optimal Freight Management for Efficiency

The management of the future heterogenous fleet is a complex decision-making problem. The heterogenous fleet is emerging as decarbonization technologies are deployed by fleets toward lowering the freight operation emissions in Medium and Heavy-duty vehicles. Traditionally, in fleets characterized by a homogeneous Diesel Internal Combustion Engine (ICE) powertrain, the process of fleet planning and operational optimization unfolds sequentially without the necessity to account for powertrain and vehicle-specific characteristics during dispatch decisions. Fleets with trucks less than 5 years old tend to maintain stable vehicle efficiency with minimal operational reliability risks for fleet managers. However, the landscape changes with the incorporation of emerging powertrain technologies, which lack extensive operational data and service experiences. This includes technologies like hybrid, Electric, Fuel Cell, or alternative fuel ICE. Operational decisions for fleets featuring heterogeneous powertrain technologies and facing limited access to alternative fueling and charging stations become intricate, requiring careful consideration and optimization at each dispatch. The difference in efficiency characteristics of emerging technologies, their range limitations, and the restricted availability of charging/alternative fueling infrastructure, coupled with sensitivity to driving conditions (e.g., EV range reduction in low temperatures) and their impact on component aging (such as batteries), become pivotal factors influencing the reliable and efficient freight transportation. To make the path toward low emission freight transportation efficient and reliable, an AI-assisted fleet management software is developed in this project to help fleet managers in optimizing both adoption of emerging powertrain decarbonization, connected and automated technologies and also operating the fleet after such technologies are deployed as schematically. Freight transportation requirements are different depending on the cargos to be shipped, customer requirements and regions of operations. This further highlights the need for software and digital solutions to tailor deployment and operation of emerging powertrain, connectivity, and automation technologies toward the specific fleet operation requirements. The fleet management optimizer was also integrated with a model of the fleet to simulate the operation of the fleet over 1 year of the baseline fleet operation (250,000+ shipments) indicating the significance of day-to-day variations on emissions and energy consumption of a freight transportation fleet. The results demonstrate ≥20% improvement in freight efficiency in terms of WTW CO2 per ton-mile of cargo shipments while all fleet operation constraints are enforced, and the cost (CapEx and OpEx) is minimized.

33 ADVANCED PROPULSION SYSTEMS↗

Smoothing HCCI heat release with vaporization-cooling-induced thermal stratification using ethanol.

Ethanol and ethanol/gasoline blends are being widely considered as alternative fuels for light-duty automotive applications. At the same time, HCCI combustion has the potential to provide high efficiency and ultra-low exhaust emissions. However, the application of HCCI is typically limited to low and moderate loads because of unacceptably high heat-release rates (HRR) at higher fueling rates. This work investigates the potential of lowering the HCCI HRR at high loads by using partial fuel stratification to increase the in-cylinder thermal stratification. This strategy is based on ethanol's high heat of vaporization combined with its true single-stage ignition characteristics. Using partial fuel stratification, the strong fuel-vaporization cooling produces thermal stratification due to variations in the amount of fuel vaporization in different parts of the combustion chamber. The low sensitivity of the autoignition reactions to variations of the local fuel concentration allows the temperature variations to govern the combustion event. This results in a sequential autoignition event from leaner and hotter zones to richer and colder zones, lowering the overall combustion rate compared to operation with a uniform fuel/air mixture. The amount of partial fuel stratification was varied by adjusting the fraction of fuel injected late to produce stratification, and also by changing the timing of the late injection. The experiments show that a combination of 60-70% premixed charge and injection of 30-40 % of the fuel at 80{sup o}CA before TDC is effective for smoothing the HRR. With CA50 held fixed, this increases the burn duration by 55% and reduces the maximum pressure-rise rate by 40%. Combustion stability remains high but engine-out NO{sub x} has to be monitored carefully. For operation with strong reduction of the peak HRR, ISNO{sub x} rises to around 0.20 g/kWh for an IMEP{sub g} of 440 kPa. The single-cylinder HCCI research engine was operated naturally aspirated without EGR at 1200 rpm, and had low residual level using a CR = 14 piston.

09 BIOMASS FUELS↗

Parallel beamlet dose calculation via beamlet contexts in a distributed multi‐GPU framework

Purpose Dose calculation is one of the most computationally intensive, yet essential tasks in the treatment planning process. With the recent interest in automatic beam orientation and arc trajectory optimization techniques, there is a great need for more efficient model‐based dose calculation algorithms that can accommodate hundreds to thousands of beam candidates at once. Foundational work has shown the translation of dose calculation algorithms to graphical processing units ( GPU s), lending to remarkable gains in processing efficiency. But these methods provide parallelization of dose for only a single beamlet, serializing the calculation of multiple beamlets and under‐utilizing the potential of modern GPU s. In this paper, the authors propose a framework enabling parallel computation of many beamlet doses using a novel beamlet context transformation and further embed this approach in a scalable network of multi‐ GPU computational nodes. Methods The proposed context‐based transformation separates beamlet‐local density and TERMA into distinct beamlet contexts that independently provide sufficient data for beamlet dose calculation. Beamlet contexts are arranged in a composite context array with dosimetric isolation, and the context array is subjected to a GPU collapsed‐cone convolution superposition procedure, producing the set of beamlet‐specific dose distributions in a single pass. Dose from each context is converted to a sparse representation for efficient storage and retrieval during treatment plan optimization. The context radius is a new parameter permitting flexibility between the speed and fidelity of the dose calculation process. A distributed manager‐worker architecture is constructed around the context‐based GPU dose calculation approach supporting an arbitrary number of worker nodes and resident GPU s. Phantom experiments were executed to verify the accuracy of the context‐based approach compared to Monte Carlo and a reference CPU ‐ CCCS implementation for single beamlets and broad beams composed by addition of beamlets. Dose for representative 4 π beam sets was calculated in lung and prostate cases to compare its efficiency with that of an existing beamlet‐sequential GPU ‐ CCCS implementation. Code profiling was also performed to evaluate the scalability of the framework across many networked GPU s. Results The dosimetric accuracy of the context‐based method displays <1.35% and 2.35% average error from the existing serialized CPU ‐ CCCS algorithm and Monte Carlo simulation for beamlet‐specific PDD s in water and slab phantoms, respectively. The context‐based method demonstrates substantial speedup of up to two orders of magnitude over the beamlet‐sequential GPU ‐ CCCS method in the tested configurations. The context‐based framework demonstrates near linear scaling in the number of distributed compute nodes and GPU s employed, indicating that it is flexible enough to meet the performance requirements of most users by simply increasing the hardware utilization. Conclusions The context‐based approach demonstrates a new expectation of performance for beamlet‐based dose calculation methods. This approach has been successful in accelerating the dose calculation process for very large‐scale treatment planning problems ‐ such as automatic 4 π IMRT beam orientation and VMAT arc trajectory selection, with hundreds of thousands of beamlets ‐ in clinically feasible timeframes. The flexibility of this framework makes it as a strong candidate for use in a variety of other very large‐scale treatment planning tasks and clinical workflows.

Neph, Ryan↗

Hydrodechlorination of 1,2-Dichloroethane on Platinum Catalysts: Insights from Reaction Kinetics Experiments, Density Functional Theory, and Microkinetic Modeling

Catalytic hydrodechlorination is a promising strategy for treating industrial 1,2-dichloroethane wastes, for which Pt and Pt-based alloy catalysts are widely used. Here, we performed a detailed mechanistic study for 1,2-dichloroethane hydrodechlorination on Pt using a synergistic approach combining density functional theory (DFT) calculations, reaction kinetics experiments, and microkinetic modeling. Using planewave DFT calculations, we evaluated the reaction energy and activation energy barrier of each elementary step involved in the reaction network on Pt(111). The calculated energetics were then incorporated into a comprehensive mean-field microkinetic model accounting for a total of 65 elementary steps. The model-predicted reaction rates were compared with the results from our reaction kinetics experiments on SiO2-supported Pt catalysts. Our results indicated that the hydrodechlorination of 1,2-dichloroethane on Pt(111) starts with a H-removal step; then, it proceeds through a sequence of alternating dechlorination and dehydrogenation steps until vinylidene (CH 2 C*) is formed; finally, CH 2 C* is hydrogenated to the final product, ethane, sequentially via vinyl (CH 2 CH*), ethylene, and ethyl (CH 3 CH 2 *) intermediates. After model parameter adjustments, we achieved good agreement between our theoretical model and experimental results; the adjustments to the calculated parameters are consistent with the typically anticipated coverage effects. Furthermore, our study offers valuable mechanistic insights, which are useful for improving catalysts for this chemistry.

1,2-dichloroethane↗

Machine Learning Accelerates Innovation in Perovskite Manufacturing Scale-up (Final Technical Report (FTR))

We propose to address the challenge of the vast parameter space associated with perovskite manufacturing optimization, by developing a machine learning (ML)-assisted optimization framework for a scalable perovskite PV manufacturing tool. This framework will be interpretable, sequential, and rapidly adaptable to upgraded systems (e.g., via transfer learning). The tool is an open-air rapid spray plasma process (RSPP) of perovskite films, which has already been established at Stanford and is a unique platform to test and deploy the proposed ML-guided framework because the RSPP technique is able to conduct optimization experiments with a high throughput, and easily adjust a wide range of process variables.

14 SOLAR ENERGY↗

SU-G-BRA-05: Application of a Feature-Based Tracking Algorithm to KV X-Ray Fluoroscopic Images Toward Marker-Less Real-Time Tumor Tracking

Purpose: To detect target position on kV X-ray fluoroscopic images using a feature-based tracking algorithm, Accelerated-KAZE (AKAZE), for markerless real-time tumor tracking (RTTT). Methods: Twelve lung cancer patients treated with RTTT on the Vero4DRT (Mitsubishi Heavy Industries, Japan, and Brainlab AG, Feldkirchen, Germany) were enrolled in this study. Respiratory tumor movement was greater than 10 mm. Three to five fiducial markers were implanted around the lung tumor transbronchially for each patient. Before beam delivery, external infrared (IR) markers and the fiducial markers were monitored for 20 to 40 s with the IR camera every 16.7 ms and with an orthogonal kV x-ray imaging subsystem every 80 or 160 ms, respectively. Target positions derived from the fiducial markers were determined on the orthogonal kV x-ray images, which were used as the ground truth in this study. Meanwhile, tracking positions were identified by AKAZE. Among a lot of feature points, AKAZE found high-quality feature points through sequential cross-check and distance-check between two consecutive images. Then, these 2D positional data were converted to the 3D positional data by a transformation matrix with a predefined calibration parameter. Root mean square error (RMSE) was calculated to evaluate the difference between 3D tracking and target positions. A total of 393 frames was analyzed. The experiment was conducted on a personal computer with 16 GB RAM, Intel Core i7-2600, 3.4 GHz processor. Results: Reproducibility of the target position during the same respiratory phase was 0.6 +/− 0.6 mm (range, 0.1–3.3 mm). Mean +/− SD of the RMSEs was 0.3 +/− 0.2 mm (range, 0.0–1.0 mm). Median computation time per frame was 179 msec (range, 154–247 msec). Conclusion: AKAZE successfully and quickly detected the target position on kV X-ray fluoroscopic images. Initial results indicate that the differences between 3D tracking and target position would be clinically acceptable.

60 APPLIED LIFE SCIENCES↗

Combined CDF and D0 upper limits on $gg\to H\to W^+W^-$ and constraints on the Higgs boson mass in fourth-generation fermion models with up to 8.2 fb$^{-1}$ of data

We combine results from searches by the CDF and D0 Collaborations for a standard model Higgs boson (H) in the processes gg {yields} H {yields} W{sup +}W{sup -} and gg {yields} H {yields} ZZ in p{bar p} collisions at the Fermilab Tevatron Collider at {radical}s = 1.96 TeV. With 8.2 fb{sup -1} of integrated luminosity analyzed at CDF and 8.1 fb{sup -1} at D0, the 95% C.L. upper limit on {sigma}(gg {yields} H) x {Beta}(H {yields} W{sup +}W{sup -}) is 1.01 pb at m{sub H} = 120 GeV, 0.40 pb at m{sub H} = 165 GeV, and 0.47 pb at m{sub H} = 200 GeV. Assuming the presence of a fourth sequential generation of fermions with large masses, we exclude at the 95% Confidence Level a standard-model-like Higgs boson with a mass between 124 and 286 GeV.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Photoproduction of $\eta$ $\pi$ pairs off nucleons and deuterons

Quasi-free photoproduction of πη-pairs has been investigated from threshold up to incident photon energies of 1.4 GeV, respectively up to photon-nucleon invariant masses up to 1.9 GeV. Total cross sections, angular distributions, invariant-mass distributions of the πη and meson-nucleon pairs, and beam helicity asymmetries have been measured for the reactions γp → pπ 0 η, γn → nπ 0 η, γp → nπ + η, and γn → pπ – η from nucleons bound inside the deuteron. For the γp initial-state data for free protons have also been analyzed. Finally, the total cross sections for quasi-free production of π 0 η pairs from nucleons bound in 3He nuclei have been investigated in view of final state interaction (FSI) effects. The experiments were performed at the tagged photon beam facility of the Mainz MAMI accelerator using an almost 4π covering electromagnetic calorimeter composed of the Crystal Ball and TAPS detectors. The shapes of all differential cross section data and the asymmetries are very similar for protons and neutrons and agree with the conjecture that the reactions are dominated by the sequential Δ *3/2 – → ηΔ(1232) → πηN decay chain, mainly with Δ(1700)3/2 – and Δ(1940)3/2 – . The ratios of the magnitude of the total cross sections also agree with this assumption. Furthermore, the absolute magnitudes of the cross sections are reduced by FSI effects with respect to free proton data.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Self-driving thin film laboratory: autonomous epitaxial atomic-layer synthesis via real-time computer vision analysis of electron diffraction

Emerging materials science platforms with the ability to make autonomous decisions on the fly are fundamentally changing the outlook and protocols for materials optimization and discovery. Because AI-driven self-navigating schemes can effectively reduce the total number of iterations needed to arrive at the "answer" (i.e. the best stochiometric composition for a desired physical property, optimum materials processing parameters, etc.) by significant margins, they have the potential to revolutionize materials and chemical manufacturing processes at large in research laboratory settings as well as in industrial plants. Here, we demonstrate a successful implementation of real-time closed-loop autonomous navigation of a multi-dimensional materials synthesis parameter space for fabricating phase-pure epitaxial films of a metastable phase of a functional oxide in a combinatorial pulsed laser deposition chamber. Sequential epitaxial growth iterations in search of the optimized recipe to stabilize the desired crystal phase were performed using frame-by-frame quantitative computer vision analysis of reflection high-energy electron diffraction (RHEED) images of the unit-cell level film being deposited. The autonomous scheme regularly resulted in > 30-fold reduction in the number of required experiments compared to a comprehensive mapping of the parameter space. The real-time workflow developed here can be readily extended to a variety of thin film synthesis platforms opening the door for self-driving atomic-level materials design as well as autonomous optimization of semiconductor manufacturing.

36 MATERIALS SCIENCE↗

Advancing reliability assessments of photovoltaic modules and materials using combined-accelerated stress testing

Previously undiscovered failure modes in photovoltaic (PV) modules continue to emerge in field installations despite passing protocols for design qualification and quality assurance. Failure to detect these modes prior to widespread use could be attributed to the limitations of present-day standard accelerated stress tests (ASTs), which are primarily designed to identify known degradation or failure modes at the time of development by applying simultaneous or sequential stress factors (usually two at most). Here, we introduce an accelerated testing method known as the combined-accelerated stress test (C-AST), which simultaneously combines multiple stress factors of the natural environment. Simultaneous combination of multiple stress factors allows for improved identification of failure modes with better ability to detect modes not known a priori. A demonstration experiment was conducted that reproduced the field-observed cracking of polyamide- (PA-) and polyvinylidene fluoride (PVDF)–based backsheet films, a failure mode that was not detected by current design qualification and quality assurance testing requirements. In this work, a two-phase testing protocol was implemented. The first cycle (“Tropical”) is a predominantly high-humidity and high-temperature test designed to replicate harsh tropical climates. The second cycle (“Multi-season”) was designed to replicate drier and more temperate conditions found in continental or desert climates. Testing was conducted on 2 × 2-cell crystalline-silicon cell miniature modules constructed with both ultraviolet (UV)–transmitting and UV-blocking encapsulants. Cracking failures were observed within a cumulative 120 days of the Tropical condition for one of the PA-based backsheets and after 84 days of Tropical cycle followed by 42 days of the Multi-season cycle for the PVDF-based backsheet, which are both consistent with failures seen in fielded modules. In addition to backsheet cracking, degradation modes were observed including solder/interconnect fatigue, various light-induced degradation modes, backsheet delamination, discoloration, corrosion, and cell cracking. The ability to simultaneously apply multiple stress factors may allow many of the test sequences within the standardized design qualification procedure to be performed using a single test setup.

14 SOLAR ENERGY↗

Characterization and Application of Superlig 620 Solid Phase Extraction Resin for Automated Process Monitoring of 90Sr

Characterization of SuperLig® 620 solid phase extraction resin was performed in order to develop an automated on-line process monitor for 90Sr. The main focus was on strontium separation from barium, with the goal of developing an automated separation process for 90Sr in high-level wastes. High-level waste contains significant 137Cs activity, of which 137mBa is of great concern as an interference to the quantification of strontium. In addition barium, yttrium and plutonium were studied as potential interferences to strontium uptake and detection. A number of complexants were studied in a series of batch Kd experiments, as SuperLig® 620 was not previously known to elute strontium in typical mineral acids. The optimal separation was found using a 2M nitric acid load solution with a strontium elution step of ~0.49M ammonium citrate and a barium elution step of ~1.8M ammonium citrate. 90Sr quantification of Hanford high-level tank waste was performed on a sequential injection analysis microfluidics system coupled to a flow-cell detector. The results of the on-line procedure are compared to standard radiochemical techniques in this paper.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Investigation of Origin of Attached Cu-Ag Droplets to Solid Particles During High-Temperature Slag/Copper/Spinel Interactions

This study investigates the origin of mechanically entrained metal droplets in liquid slag due to their interaction with solid spinel particles. Two possible mechanisms were proposed previously: separately formed droplets and spinel particles get attached to each other due to agitation of the slag and metal phases; or the spinel particles form by a chemical reaction together with a new droplet or alongside a droplet that was already present in the system. In this study, an inert tracer element was added to the metallic phase in adapted sessile drop experiments. For this purpose, Cu-Ag alloys, with various Ag-contents, were produced. The results showed that the small entrained metal droplets within the slag droplet contained Ag, but in very low amounts with respect to the amount of Ag in the Cu-Ag alloy. This indicates that the entrained metal droplets are formed due to a sequential combination of the two origins: first, very small metal droplets are dispersed in the slag drop, due to the emulsification process. Then, these metal droplets are nucleation sites for the Cu-spinel reactive formation.

36 MATERIALS SCIENCE↗

Spotlight: efficient automated global optimization in rietveld analysis of diffraction data

Performing reliable Rietveld analysis on tens or hundreds of powder diffraction datasets from parametric or time-resolved experiments often poses a bottleneck in extracting meaningful results from the data. While automated analysis of data has recently been demonstrated, high temperature annealing studies, during which phase transformations occur and lattice parameters may change due to repartitioning of elements, are prime examples where automation by a simple phase identification from a database of room temperature structures or automation by sequential refinements is likely to fail. To enable reliable, efficient, automated Rietveld analysis, we present a Python package named Spotlight , building on established Rietveld packages such as MAUD, GSAS , or GSAS-II , which extends the refinement of best fit parameters to a global optimization using an ensemble of optimizers leveraging hierarchical parallel execution on high-performance computing clusters. Spotlight further enables the efficient design of refinement plans through the iterative automated machine-learning of a surrogate for the refinement on which the global optimizations are performed until results from the surrogate converge to the response surface data. We demonstrate Spotlight with the analysis of uranium molybdenum and Ti–6Al–4V datasets, as well as in two open-source tutorials analyzing aluminium oxide and lead sulphate.

36 MATERIALS SCIENCE↗

Sequential Bayesian Experimental Design for Calibration of Expensive Simulation Models

Simulation models of critical systems often have parameters that need to be calibrated using observed data. For expensive simulation models, calibration is done using an emulator of the simulation model built on simulation output at different parameter settings. Using intelligent and adaptive selection of parameters to build the emulator can drastically improve the efficiency of the calibration process. The article proposes a sequential framework with a novel criterion for parameter selection that targets learning the posterior density of the parameters. The emergent behavior from this criterion is that exploration happens by selecting parameters in uncertain posterior regions while simultaneously exploitation happens by selecting parameters in regions of high posterior density. Furthermore, the advantages of the proposed method are illustrated using several simulation experiments and a nuclear physics reaction model.

97 MATHEMATICS AND COMPUTING↗