Search NASA⌕ Search

SEARCH · Search NASA

Results for “random processes”

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 1,621 records · Page 90

LOCAL GROUP DWARF SPHEROIDALS: CORRELATED DEVIATIONS FROM THE BARYONIC TULLY-FISHER RELATION

Local Group dwarf spheroidal satellite galaxies are the faintest extragalactic stellar systems known. We examine recent data for these objects in the plane of the Baryonic Tully-Fisher Relation (BTFR). While some dwarf spheroidals adhere to the BTFR, others deviate substantially. We examine the residuals from the BTFR and find that they are not random. The residuals correlate with luminosity, size, metallicity, ellipticity, and susceptibility of the dwarfs to tidal disruption in the sense that fainter, more elliptical, and tidally more susceptible dwarfs deviate farther from the BTFR. These correlations disfavor stochastic processes and suggest a role for tidal effects. We identify a test to distinguish between {Lambda}CDM and MOND based on the orbits of the dwarf satellites of the Milky Way and how stars are lost from them.

79 ASTRONOMY AND ASTROPHYSICS↗

Learning to Unknot

We introduce natural language processing into the study of knot theory, as made natural by the braid word representation of knots. We study the UNKNOT problem of determining whether or not a given knot is the unknot. After describing an algorithm to randomly generate N-crossing braids and their knot closures and discussing the induced prior on the distribution of knots, we apply binary classification to the UNKNOT decision problem. We find that the Reformer and shared-QK Transformer network architectures outperform fully-connected networks, though all perform at 95% accuracy. Perhaps surprisingly, we find that accuracy increases with the length of the braid word, and that the networks learn a direct correlation between the confidence of their predictions and the degree of the Jones polynomial. Finally, we utilize reinforcement learning (RL) to find sequences of Markov moves and braid relations that simplify knots and can identify unknots by explicitly giving the sequence of unknotting actions. Trust region policy optimization (TRPO) performs consistently well, reducing 80% of the unknots with up to 96 crossings we tested to the empty braid word, and thoroughly outperformed other RL algorithms and random walkers. Studying these actions, we find that braid relations are more useful in simplifying to the unknot than one of the Markov moves.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

An efficient method to propagate model uncertainty when inverting seismic data for time domain seismic moment tensors

SUMMARY We present a computationally efficient method to approximately propagate uncertainty when linearly inverting seismic data for point source, time variable moment tensor components. The method is based on the assumption that the data residual, given by the difference between the observed seismic data and the data predicated by a linear inversion, contains the effects of both data and model uncertainty. Our method uses a distribution of data residuals, added directly to the data, in a pseudo-Monte Carlo scheme. Using the assumption that the data residual is a stochastic process, we use the well-known Karhunen–Loève (KL) theorem to construct a distribution of data residuals, where the required basis functions are constructed using Fourier series. The Fourier series are scaled by a product of a random variable and the real-valued spectral amplitudes of the original data residual’s spectrum. Thus, the Fourier series and spectral amplitudes are eigenfunction-eigenvalue pairs used in the KL-based construction of data residual distribution. Using tests with synthetic data, we show that our method compares closely with a Finite Difference Monte Carlo (FDMC) method that we presented previously. More importantly, the method presented here is computationally several orders of magnitude faster than our previous FDMC method, and requires no a priori assumptions of model and/or data uncertainty.

Poppeliers, Christian (ORCID:0000000159526849)↗

A CMOS Compatible, Forming Free TaO x ReRAM

Resistive random access memory (ReRAM) has become a promising candidate for next-generation high-performance non-volatile memory that operates by electrically tuning resistance states via modulating vacancy concentrations. Here, we demonstrate a wafer-scale process for resistive switching in tantalum oxide that is completely CMOS compatible. The resulting devices are forming-free and with greater than 1x10 5 cycle endurance.

97 MATHEMATICS AND COMPUTING↗

Codon2Vec v1.0

Background: Codon2Vec is an embedding neural network that predicts 'high' or 'low' gene expression directly from the protein-coding sequences. Embedding neural networks are commonly used for natural language processing (NLP) applications. Analogous to how an English sentence is a string of words, a gene can be thought of as a string of codons. Similar to how NLP neural networks model English sentences as a non-random sequence of words, we considered a coding sequence as a non-random non-overlapping array of codons (k-mers of length = 3). Value Proposition: - Codon2Vec achieved a high median AUC-ROC score of 83.8% when trained and applied to transcriptomic data from 300 fungal species - Unlike Codo2Vec, conventional methods predicting for expression based on codon usage rely on a priori knowledge of optimal codons or a set of reference genes. - Unlike Codon2vec, these methods do not account for the effect of codon order on gene expression. - Codon2Vec neural network bypasses the need for artisanal feature selection step that is necessary for traditional machine learning models.

Wint, Rhondene↗

Development of Automated Software Program for the Analysis of Alzheimer's Disease Scans: Evaluation of PET imaging with PBR-111 in HV and AD subjects

Study goal: A Phase 1 evaluation of the kinetics, clearance and cerebral distribution of one novel peripheral benzodiazepine receptors(PBR)positron emission tomography (PET) imaging agent, 18F-PBR-111 following intravenous administration in healthy volunteers and Alzheimer's disease (AD) patients. Short title: Evaluation of PET imaging with PBR-111 in HV and AD subjects Proof of Mechanism. Primary Objective: To evaluate the cerebral distribution of PBR-111 positron emission tomography (PET) for detection/exclusion of microglial activation in patients with Alzheimer's disease subjects compared to healthy volunteers. Secondary objectives: - To assess the dynamic uptake and washout of [18F]PBR-111, a potential imaging bio-marker for inflammatory changes in brain, using positron emission tomography in subjects with Alzheimer's disease (AD) and healthy volunteers (HV). - To perform blood metabolite characterization of [18F]PBR-111 in subjects with AD and HV to determine the nature of metabolites in assessment of [18F]-PBR-111 as a PET brain imaging agent. Name of radioactive drug substance: PBR-111 Dose(s): The applied PBR-111 radioactive dose will be up to 5.0 mCi, diluted in a maximum of 10 ml of saline. The radioligand will be administered as a slow intravenous bolus injection (i.e., 6 sec/ml) into a large vein (e.g., antecubital vein). Route of administration: Intravenous injection Duration of treatment: Single administration of a diagnostic agent Indication: PBR-111 positron emission tomography (PET) imaging has the potential to detect microglial activation. In the presence of PBR-111 uptake (representative of microglial activation), inflammation in the brain can be detected. Diagnosis and main criteria for inclusion: Study participants will be HVs and patients diagnosed with probable AD. HVs must be 18 years of age (at least four subjects 50 years of age) and have no evidence of cognitive impairment or other neurologic disease by medical history. The lack of cognitive impairment will also be based on a Clinical Dementia Rating (CDR) of 0. Patients with probable AD must be 50 years of age and must fulfill the National Institute of Neurological and Communicative Disorders and Stroke, Alzheimer's Disease and Related Disorders Association [NINCDS-ADRDA] criteria for probable AD. The CDR score must be 1.0 and 2.0 and have a modified Hachinski of 4. All HVs and all patients with probable AD must be able to comply with all study procedures. Study design: This is a Phase 1, open-label, single-center, non-randomized single dose study to assess the kinetics, clearance and cerebral distribution of PBR-111 PET imaging in detecting microglial activation in the brain in patients with probable AD compared to HVs. All aspects related to image acquisition, processing, and visual as well as quantitative evaluation will be developed, optimized, and validated (where required). Each subject will be required to visit the study center during the screening phase and on the PBR-111 PET imaging day (baseline). A telephone follow-up visit will be performed 7 days (± 3 days) after PBR-111 PET administration. At the screening visit, each subject (or caregiver in the case of AD subjects) will be asked to provide written informed consent or assent. During the screening phase (maximum duration of 60 days) subject medical, neurological, and surgical history, clinical assessments, and a neuro-psychiatric evaluation will be performed on all eligible subjects. Subjects will be allowed to leave the center after all evaluations have been completed. During this period an MRI of the brain will be performed during the screening period. If an MRI of the brain has been performed within six months of the imaging visit using the methods described in the protocol, and there has been no medically significant events in the interim, the previous MRI may be used. During the PBR-111 PET imaging day, all subjects will receive a single intravenous injection of PBR-111 and scanning will be performed over a 3.5 hour period. Each subject will have a telephone follow-up 7 days (± 3 days) thereafter to assess for adverse events. Methodology: - Assessments to provide clinical characterization of the AD subjects will be performed. - After administration of PBR-111, images will be generated with state-of-the-art PET imaging. Images will be assessed quantitatively for the presence of microglial activation by a nuclear physician blinded to clinical data. - Total radioactivity and estimation of the fraction of radioactivity associated to the un-metabolized tracer will be determined. In addition, the metabolite patterns of PBR-111 are determined in venous plasma and arterial samples based on high-performance liquid chromatography (HPLC) analyses. - Arterial sampling will be acquired in the initial two AD and two HV subjects and modeling will be assessed to determine if additional arterial sampling is necessary.

97 MATHEMATICS AND COMPUTING↗

Low Cost Corrugated Substrates for High Efficiency OLEDs (Final Report)

The external quantum efficiency (EQE) of a planar OLED is limited to 25%. Around 75% of the light is trapped in the substrate, waveguided and SPP modes. In this project, we demonstrated that nano-structures can significantly improve the EQE of OLEDs. The objective is to improve the EQE to 70% using corrugated substrate, low index layer incorporation and micro lens arrays. We conducted optical study of the operating devices using ARES measurement, and used the result to guide the design of corrugated substrates as well as the OLED stack for better performance. During the first year of the program, we have demonstrated over 60% EQE in a corrugated green OLED with half-ball lens for light extraction. By comparing, the optical characteristics of OLEDs fabricated on single and poly crystal corrugated structures, we confirm that the enhancement using corrugated substrates comes from waveguided and SPP modes extraction. To push the device EQE to 70%, further optimization of the corrugated substrate is needed. The extraction of waveguided and SPP modes are influenced by the corrugation pitch, depth and profile. Therefore, we developed a series of processes to precisely control the corrugation morphology. Through comparison of different corrugated substrates, we observed that while an abrupt surface profile results in stronger mode extraction, it also introduces strong leakage current; while a smooth profile are less efficient in mode extraction, it preserves the quality of an OLED. Based on these observations, we conclude that a single crystal hole corrugation results in optimum EQE % and low leakage current when planarized with PEDOT. To improve the extraction of waveguided mode, we also introduce a low index contrast layer (ICL) to increase the index contrast between corrugated ITO and glass interface. We optimized our recipe for the index contrast layer processing and reduced its refractive index from 1.50 to 1.38. As a result, the optimized device shows 47.6% EQE, which is 1.64 times higher than the reference planar device. By extracting the substrate mode with a hemispherical lens, we obtained an EQE of 71.6%, which is a 2.46 times enhancement. The cross-section SEM image reveals the high EQE is due to the formation of an air gap at the interface of the ICL, resulting in an increase of the index contrast to Δn ~ 1. We use angle-resolved electroluminescence spectroscopy (ARES) to demonstrate that the ICL improves the diffraction efficiency at the ITO/glass interface, extracts TM, TE waveguide and SPP mode more efficiently, and leads to the observed high efficiency. We also address the directional emission issue with the single crystalline microlens arrays (MLAs) by fabricating polycrystalline MLAs. The process uses colloidal assembly method to create a mono-layer of silica spheres, then uses PDMS stamps to replicate the pattern for multi-use imprinting. The MLAs are close-packed in short range but have random orientations in the long range. This avoids stronger extraction along the optical axes of single crystal MLAs and prevents the directional emission issue. We also demonstrate that the color shift induced by the corrugation can be minimized using MLAs. In this scenario, the MLAs work as diffuser to randomize the directional emission from extracted waveguide and SPP modes.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Machine Learning Analysis of Hydrologic Exchange Flows and Transit Time Distributions in a Large Regulated River

Hydrologic exchange between river channels and adjacent subsurface environments is a key process that influences water quality and ecosystem function in river corridors. High-resolution numerical models were often used to resolve the spatial and temporal variations of exchange flows, which are computationally expensive. In this study, we adopt Random Forest (RF) and Extreme Gradient Boosting (XGB) approaches for deriving reduced order models of hydrologic exchange flows and associated transit time distributions, with integrated field observations (e.g., bathymetry) and hydrodynamic simulation data (e.g., river velocity, depth). The setup allows an improved understanding of the influences of various physical, spatial, and temporal factors on the hydrologic exchange flows and transit times. The predictors also contain those derived using hybrid clustering, leveraging our previous work on river corridor system hydromorphic classification. The machine learning-based predictive models are developed and validated along the Columbia River Corridor, and the results show that the top parameters are the thickness of the top geological formation layer, the flow regime, river velocity, and river depth; the RF and XGB models can achieve 70% to 80% accuracy and therefore are effective alternatives to the computational demanding numerical models of exchange flows and transit time distributions. Each machine learning model with its favorable configuration and setup have been evaluated. The transferability of the models to other river reaches and larger scales, which mostly depends on data availability, is also discussed.

97 MATHEMATICS AND COMPUTING↗

Micro-tidal disruption events by stellar compact objects and the production of ultra-long GRBs

We explore full/partial tidal disruption events (TDEs) of stars/planets by stellar compact objects (black holes (BHs) or neutron stars (NSs)), which we term micro-TDEs. Disruption of a star/planet with mass M {sub ⋆} may lead to the formation of a debris disk around the BH/NS. Efficient accretion of a fraction (f{sub acc}=0.1 of the debris may then give rise to bright, energetic, long (10{sup 3}–10{sup 4} s), X-ray/gamma-ray flares, with total energies of up to (f{sub acc}/0.1)×10{sup 52} (M{sub ⋆}/0.6 M{sub ⊙}) erg, possibly resembling ultra-long gamma-ray bursts (GRBs)/X-ray flashes (XRFs). The energy of such flares depends on the poorly constrained accretion processes. Significantly fainter flares might be produced if most of the disk mass is blown away through strong outflows. We suggest three dynamical origins for such disruptions. In the first, a star/planet is tidally disrupted following a close random encounter with a BH/NS in a dense cluster. We estimate the BH (NS) micro-TDE rates from this scenario to be a few ×10{sup −6} (a few ×10{sup −7}) yr{sup −1} per Milky Way galaxy. Another scenario involves the interaction of wide companions due to perturbations by stars in the field, likely producing comparable but lower rates. Finally, a third scenario involves a BH/NS that gains a natal velocity kick at birth, leading to a close encounter with a binary companion and the tidal disruption of that companion. Such events could be associated with a supernova, or even with a preceding GRB/XRF event, and would likely occur hours to days after the prompt explosion; the rates of such events could be larger than those obtained from the other scenarios, depending on the preceding complex binary stellar evolution.

79 ASTRONOMY AND ASTROPHYSICS↗

Chemical Doppelgangers in GALAH DR3: The Distinguishing Power of Neutron-capture Elements among Milky Way Disk Stars

The observed chemical diversity of Milky Way stars places important constraints on Galactic chemical evolution and the mixing processes that operate within the interstellar medium. Recent works have found that the chemical diversity of disk stars is low. For example, the Apache Point Observatory Galactic Evolution Experiment (APOGEE) "chemical doppelganger rate," or the rate at which random pairs of field stars appear as chemically similar as stars born together, is high, and the chemical distributions of APOGEE stars in some Galactic populations are well-described by two-dimensional models. However, limited attention has been paid to the heavy elements (Z > 30) in this context. In this work, we probe the potential for neutron-capture elements to enhance the chemical diversity of stars by determining their effect on the chemical doppelganger rate. We measure the doppelganger rate in GALactic Archaeology with HERMES DR3, with abundances rederived using The Cannon, and find that considering the neutron-capture elements decreases the doppelganger rate from ~2.2% to 0.4%, nearly a factor of 6, for stars with –0.1 < [Fe/H] < 0.1. While chemical similarity correlates with similarity in age and dynamics, including neutron-capture elements does not appear to select stars that are more similar in these characteristics. Our results highlight that the neutron-capture elements contain information that is distinct from that of the lighter elements and thus add at least one dimension to Milky Way abundance space. This work illustrates the importance of considering the neutron-capture elements when chemically characterizing stars and motivates ongoing work to improve their atomic data and measurements in spectroscopic surveys.

79 ASTRONOMY AND ASTROPHYSICS↗

Ternary PtSnRh–SnO2 Nanoclusters: Synthesis and Electroactivity for Ethanol Oxidation Fuel Cell Reaction

Carbon supported ternary Pt{sub 52}Sn{sub (36-x)}Rh{sub 12}-Sn{sub x}O{sub 2x} electrocatalysts with the average diameter of 2.8 {+-} 0.5 nm were synthesized using a Polyol process followed by thermal treatment. Several techniques including high resolution transmission electron microscopy (HRTEM), X-ray diffraction (XRD) and extended X-ray absorption fine structure (EXAFS) were used to identify the coexistence of homogeneously distributed Pt/Sn/Rh random alloy and non-alloyed SnO{sub 2} throughout the catalyst. The Pt{sub 52}Sn{sub (36-x)}Rh{sub 12}-Sn{sub x}O{sub 2x} catalyst showed a superior long-term activity and stability towards ethanol oxidation than the commercial Pt catalyst. Our data of ternary Pt/Sn/Rh catalysts with different chemical compositions and crystalline structures also indicated that the superior performance of Pt{sub 52}Sn{sub (36-x)}Rh{sub 12}-Sn{sub x}O{sub 2x} might result from the electronic effect of the Pt/Sn/Rh random alloy.

30 DIRECT ENERGY CONVERSION↗

Randomized Algorithms for Scientific Computing (RASC)

Randomized algorithms have propelled advances in artificial intelligence (AI) and represent a foundational research area in advancing AI for Science. Future advancements in DOE Office of Science priority areas such as climate science, astrophysics, fusion, advanced materials, combustion, and quantum computing all require randomized algorithms for surmounting challenges of complexity, robustness, and scalability. Advances in data collection and numerical simulation have changed the dynamics of scientific research and motivate the need for randomized algorithms. For instance, advances in imaging technologies such as X-ray ptychography, electron microscopy, electron energy loss spectroscopy, or adaptive optics lattice light-sheet microscopy collect hyperspectral imaging and scattering data in terabytes, at breakneck speed enabled by state-of-the-art detectors. The data collection is exceptionally fast compared with its analysis. Likewise, advances in high-performance architectures have made exascale computing a reality and changed the economies of scientific computing in the process. Floating-point operations that create data are essentially free in comparison with data movement. Thus far, most approaches have focused on creating faster hardware. Ironically, this faster hardware has exacerbated the problem by making data still easier to create. Under such an onslaught, scientists often resort to heuristic deterministic sampling schemes (e.g., low-precision arithmetic, sampling every nth element) and sacrifice potentially valuable accuracy. Dramatically better results can be achieved via randomized algorithms, reducing the data size as much as or more than naive deterministic subsampling can achieve, while retaining the high accuracy of computing on the full data set. By randomized algorithms we mean those algorithms that employ some form of randomness in internal algorithmic decisions to accelerate time to solution, increase scalability, or improve reliability. Examples include matrix sketching for solving large-scale least-squares problems (see Figure 1) and stochastic gradient descent for training machine learning models. We are not recommending heuristic methods but rather randomized algorithms that have certificates of correctness and probabilistic guarantees of optimality and near-optimality. Such approaches can be useful beyond acceleration, for example, in understanding how to avoid measure zero worst-case scenarios that plague methods such as QR matrix factorization.

97 MATHEMATICS AND COMPUTING↗

Room-temperature fabrication of a delafossite CuCrO 2 hole transport layer for perovskite solar cells

Delafossite oxides are promising hole transport layer (HTL) candidates for perovskite solar cells, due to their wide band gap, favorable energy band alignment relative to the perovskite absorber and simplicity of processing. In this paper, we investigate the properties of CuCrO2 (CCO) delafossite films using an integration of experimental and computational techniques. Phase-pure CCO films are deposited at room temperature by spin-casting suspensions of hydrothermally-synthesized nanoparticles, for use in a glass/ITO/CCO/CH3NH3PbI3/C60/BCP/Ag device structure. Although density functional theory (DFT) calculations predict an elevated hole effective mass along certain crystallographic directions, the nearly isotropic shape and small size of the CCO nanoparticles preserves transport properties within the films by randomizing particle orientation, preventing these unfavorable directions from dominating the film conductivity. Experimental measurements confirm that the CCO films display appropriate optical and electrical properties for use as an HTL, in good agreement with the DFT calculations. Solar cells made using these films exhibit stabilized power conversion efficiencies exceeding 14%, with only minor hysteresis in the current–voltage characteristics.

Dunlap-Shohl, Wiley A.↗

ISOLOK VALVE ACCEPTANCE TESTING FOR DWPF SME SAMPLING PROCESS

Evaluation of the Defense Waste Processing Facility (DWPF) Chemical Process Cell (CPC) cycle time identified several opportunities to improve the CPC processing time. Of the opportunities, a focus area related to optimizing the equipment and efficiency of the sample turnaround time for DWPF Analytical Laboratory was identified. The Mechanical Systems & Custom Equipment Development (MS&CED) Section of the Savannah River National Laboratory (SRNL) evaluated the possibility of using an Isolok{reg_sign} sampling valve as an alternative to the Hydragard{reg_sign} valve for taking process samples. Previous viability testing was conducted with favorable results using the Isolok sampler and reported in SRNL-STI-2010-00749 (1). This task has the potential to improve operability, reduce maintenance time and decrease CPC cycle time. This report summarizes the results from acceptance testing which was requested in Task Technical Request (TTR) HLW-DWPF-TTR-2010-0036 (2) and which was conducted as outlined in Task Technical and Quality Assurance Plan (TTQAP) SRNL-RP-2011-00145 (3). The Isolok to be tested is the same model which was tested, qualified, and installed in the Sludge Receipt Adjustment Tank (SRAT) sample system. RW-0333P QA requirements apply to this task. This task was to qualify the Isolok sampler for use in the DWPF Slurry Mix Evaporator (SME) sampling process. The Hydragard, which is the current baseline sampling method, was used for comparison to the Isolok sampling data. The Isolok sampler is an air powered grab sampler used to 'pull' a sample volume from a process line. The operation of the sampler is shown in Figure 1. The image on the left shows the Isolok's spool extended into the process line and the image on the right shows the sampler retracted and then dispensing the liquid into the sampling container. To determine tank homogeneity, a Coliwasa sampler was used to grab samples at a high and low location within the mixing tank. Data from the two locations were compared to determine if the contents of the tank were well mixed. The Coliwasa sampler is a tube with a stopper at the bottom and is designed to obtain grab samples from specific locations within the drum contents. A position paper (4) was issued to address the prototypic flow loop issues and simulant selections. A statistically designed plan (5) was issued to address the total number of samples each sampler needed to pull, to provide the random order in which samples were pulled and to group samples for elemental analysis. The TTR required that the Isolok sampler perform as well as the Hydragard sampler during these tests to ensure the acceptability of the Isolok sampler for use in the DWPF sampling cells. Procedure No.L9.4-5015 was used to document the sample parameters and process steps. Completed procedures are located in R&D Engineering job folder 23269.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

XRCT characterisation of Ti particles inside porous Al{sub 2}O{sub 3}

Computed X-ray tomography was used to characterise distribution and sphericity of Ti granules within highly porous (> 35%) Al{sub 2}O{sub 3} powder compacts, as they are key parameters for a successful infiltration by Fe-based alloys. Setting of reconstruction constraints, image editing as well as data processing are the most challenging parts of computed X-ray tomography and principal sources of errors that bias the generated data. Thus, corrective measures have to be applied and the reliability of generated data has to be proved with respect to statistical, stereological and volumetric aspects. Combining an adapted Interface Particle Treatment Algorithm with the Marching Cube Method, Equilibrium Random State Model, cluster splitting and conventional laser diffraction measurements a significant improvement of the three-dimensional reconstructed data was achieved. This study points out the need of the applied algorithms for the proof and improvement of generated data by computed X-ray tomography and gives a short survey of methods that can be applied.

36 MATERIALS SCIENCE↗

A Comparison of Forecast Error Generators for Modeling Wind and Load Uncertainty

This paper presents four algorithms to generate random forecast error time series, including a truncated-normal distribution model, a state-space based Markov model, a seasonal autoregressive moving average (ARMA) model, and a stochastic-optimization based model. The error time series are used to create real-time (RT), hour-ahead (HA), and day-ahead (DA) wind and load forecast time series that statistically match historically observed forecasting data sets, used for variable generation integration studies. A comparison is made using historical DA load forecast and actual load values to generate new sets of DA forecasts with similar stoical forecast error characteristics. This paper discusses and compares the capabilities of each algorithm to preserve the characteristics of the historical forecast data sets.

Lu, Ning↗

Informative Path Planning for Mapping Radiation

This project involves making measurements of contaminated areas using portable radiation detectors and small robots. Tasks involve operating detectors and small robots, combining data acquisition for the two systems, and automating the systems. The end result is a radiation map of contamination areas. Localization of radioactive sources and mapping of contamination distribution is essential when responding to emergencies, or when dealing with cleanup efforts. Radiation maps are used for situational awareness and for establishing a safe perimeter. However, it may be hazardous or impossible for humans to safely enter the site and survey the area. Informative Path Planning (IPP) is used to design a path for a robotic sensor platform to gather the most information about the radiation distribution while operating under the set of constraints given by the dynamics of the robot and other requirements such as minimizing the overall mission time. The key challenge for IPP is the tight coupling in multiple layers of decisions: Selection of locations to visit; Allocation of sensors to those locations; Processing of information gathered. Given a robot equipped with a basic gamma detector and Lidar sensor, how can we efficiently explore an unknown environment in such a way that we can satisfy the robot's constraints while simultaneously maximizing the accuracy of the map and minimizing the total runtime? This work presents a systematic procedure for IPP and environmental mapping that: Selects the best locations at which to sample; Finds the best path for robots; Estimates the measurement value at unknown locations using a Gaussian Process regression framework. Given a set of points where radiation measurements were taken, we are utilizing a Gaussian Process Regression with a Matern kernel to reconstruct the shape of the underlying ground truth. Mapping is done via the Robot Operating System's OpenSlam Gmapping package, which creates a 2D occupancy grid map. Simulated trials were run with a radiation source placed at random locations, and allowing the robot to select and move towards 20 samples at separate locations, as well as sampling along the path while moving. The simulated sensor model assumed an efficiency of 0.3, collection time of 1 second. Various weighting schemes were tested and compared. Quantitative evaluations often favored exploitation-heavy weighting schemes where μ was emphasized, with the Root Mean Squared Error (RMSE) quickly tending towards 2, whereas qualitative comparisons would favor exploration-biased schemes with greater weight on σ. The difference comes from operators often favoring knowing the general location of a source, rather than the exact properties of a source. Weighting of the path-length π often reduced the quality of the regression as less exploration occurred, however this may be deemed a worthwhile trade-off under certain circumstances, such as with sensitive robots. Successfully developed a methodology for informative path-planning for a robot to efficiently gather information and create a map of radiation. In the future, we would like to focus on: Hardware trials with integration of the robot platform and the IPP software; More refined sensor and environmental modeling; Multiple robot platforms.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Statistical analysis of spatial distribution of pores in metal additive manufacturing

We report mechanical properties of additively manufactured parts are sensitive to the presence of pores form during the manufacturing process. The impact of pores on the mechanical performance has been investigated extensively with respect to different parameters, such as pore volume fraction, shape, and size. However, statistical investigations focusing on the relationships between the spatial distribution of pores and process parameters; and consequently, the performance of the manufactured parts are scattered and limited. Also, these sparse investigations usually suffer from an ambiguous definition of terminologies. For instance, the required criteria to consider a point pattern as complete spatial randomness (CSR) are generally not clarified. Moreover, no numerical formalism is yet developed to show how much the observed spatial results are statistically significant. To address these shortcomings, the statistical definition of CSR in a point pattern and the procedure to quantitatively determine the deviation of a pattern from CSR were discussed. The explained statistical approach was used to investigate the effect of scanning speed parameter on the spatial distribution of gas pores in laser powder bed fusion manufactured stainless steel parts. Furthermore, and to highlight the impact of the spatial distribution of pores on mechanical properties, fatigue performances of parts with clustered and randomly distributed pores were simulated by finite element analysis. It is shown that by reducing the scanning speed, the spatial distribution of gas pores deviated more from CSR, and correspondingly fatigue performance deteriorated.

36 MATERIALS SCIENCE↗