Search NASA⌕ Search

SEARCH · Search NASA

Results for “sequential sampling”

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 217 records · Page 12
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Particle Filter Based Inference Testing

The primary intent of PAR-FIT (Particle Filter based Inference Testing) is to provide hard inductive evidence that a machine learning model is capable and proven for an individual test input. By examining training data used to form the underlying model functional correlation, an estimate of the reliability that a model will make the correct prediction can be made. The Sequential Probability Ratio Test is used to derive a qualitative evaluation for reliability based on hypothesis testing. The PAR-FIT framework achieves this by implementing a particle filter and the sequential probability ratio test algorithms on the machine learning model training data to determine relevancy of new individual test samples to the training dataset. The kernel function evaluates the local proximity and density of training data used to derive a prediction outcome. Particles are used to probabilistically determine which training data to evaluate for proximity. For test samples that are within a close proximity to and surrounded by multiple training data points, the evaluated reliability of the prediction is high. For test samples that are anomalies not represented by the training dataset, in low density data clusters, or are far from existing data points, the evaluated reliability is low as insufficient training evidence exists to suggest the model is capable of making the correct prediction. Sequential Probability Ratio Test is further used to determine when a hypothesis on whether a signal can be rejected or accepted for use. The ratio test collects sequence information from the particle filter to test whether the signal is anomalous or normal via hypothesis testing of the underlying distributions.

Chen, Edward [Idaho National Laboratory (INL), Ida↗
DOE OSTI2025

Hyperspectral imaging for real-time waste materials characterization and recovery using endmember extraction and abundance detection

Hyperspectral imaging, combined with advanced spectral unmixing techniques and artificial intelligence, offers a powerful solution for improving material identification and classification. Here, this study evaluates the effectiveness of the pixel purity index and the sequential maximum angle convex cone algorithms in extracting and validating spectral signatures from pure samples of paper components (cellulose and lignin) and plastic (polypropylene). Principal-component analysis showed that both algorithms captured nearly all relevant variance for the tested materials. Spectral signatures were compared using the spectral angle mapper, revealing high similarity in the short-wave infrared region and greater variability in the visible near-infrared range. The methodology was then applied to a disposable coffee cup to detect and quantify mixed materials, accurately estimating material abundance and object area with less than 1% error. This approach enhances material classification, supporting product verification, quality control, and automated sorting for sustainable waste management and resource recovery.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Physical, Chemical, and Mineralogical Characterizations of MSWI Ash Product and Recommendations for Downstream Processing

The primary objectives of this project are to (1) systematically characterize MSWI ash, and (2) based on characterization findings, design preliminary flowsheets for downstream processing. To achieve these objectives, a total of ten tasks were completed, including sample collection, physical separation tests, liberation tests, synthetic MSWI ash preparation, elemental composition analysis, sequential chemical extraction, mineralogical characterization, pozzolanic activity characterization, thermal stability characterization, processing flowsheet design, TEA and T2M, and project performance reporting. Many useful findings and conclusions were obtained from the exhaustive efforts of this project from several different aspects, including: a) Valuable Metals in MSWI Ash: MSWI ash contains a diverse array of valuable metals. Based on potential recoverable values, the most valuable metals present in MSWI ash include Fe, Ti, Mn, Cu, Zn, V, Co, Ni, Sr, Sn, Ag, Mo, and Sc. Some of these metals have been identified as critical minerals by DOE and DOI, suggesting that MSWI is a promising feedstock for critical mineral recovery. Noticeable graphical and seasonable variations in the valuable metal content of MSWI ash were observed. Nevertheless, it was challenging to discern any clear, definitive patterns for conclusions from those observations. Compared with bottom ash, fly ash contains more volatile metals, such as Zn and Sn, but less nonvolatile metals, such as Fe, Mn, Cu, Zn, Co, and Ni. Mineralogical analyses showed that MSWI ash contains a substantial amount of calcium minerals, such as portlandite, lime, gypsum, and calcite. In addition, it was found that different types of valuable metals often exist in the same particles. b) Physical Separation of MSWI Ash: Both dry sieving and wet sieving were performed on MSWI ash. A notable disparity in the size distribution of the same material was observed when using the two different sieving methods. The disparity is due to the agglomeration of small particles. For the valuable metals investigated, no significant enrichment in a specific size fraction was observed, suggesting that it is challenging to preconcentrate the valuable metals through size fractionation. Due to the presence of ferromagnetic materials, such as Fe, most of the materials reported to the magnetic products obtained by dry magnetic separation. However, the enrichment effect is minimal due to the existence of particle agglomerates. Density separation at a cut-off density of 2.7 SG or higher led to noticeable enrichment of selected valuable metals, particularly Ti. The unburned carbon present in MSWI ash was effectively removed by flotation using diesel as the collector. A novel reagent scheme, Na2S plus cationic collectors, that can efficiently beneficiate nonferrous metals plus Co was developed. c) Liberation Tests: The particle size of MSWI ash was effectively reduced by grinding, and as a result, the encapsulated valuable metal particles (if any) were liberated to a certain degree. However, particle size reductions did not noticeably enhance the beneficiation performance using the physical separation methods, primarily due to the inefficiency of these methods in processing fine particles and/or a possibility that insufficient liberation is not a limiting factor for achieving satisfactory physical separation performance. Valuable metals were classified into water leachable, ion-exchangeable, acid soluble, reducible, oxidable, and insoluble forms. It was found that the distributions in the different categories, i.e., the occurrence modes of the valuable metals, were not affected by the particle size. d) Leaching Characteristics of Metals from MSWI Ash: Most of the valuable metals were extracted from the fly ash samples when using 1 M HCl or HNO3 as the lixiviant. The leaching reaction is a very fast process, which can reach equilibrium within the first 5 min. The releasing of Co, Ni and Ag are sensitive to leaching temperature, a higher recovery value could be obtained when using relatively higher leaching temperatures. The leachability of the valuable metals present in MSWI bottom ash is relatively lower than that of fly ash. Leaching recoveries increased with elevations in the acid concentration. Relatively high leaching recoveries were obtained for REEs, Mn, Co, Ni, Cu, and Zn using 1 M HCl or HNO3 as the lixiviant. Elevations in the reaction temperature noticeably increased the leachability of the valuable metals, whereas the leachability was barely influenced by oxidizing and reducing agents. Similar to fly ash, leaching valuable metals from bottom ash is a rapid process, with most of the leaching reaction completed within the first 5 minutes. e) Combusted iPhones: The original structure of iPhones was remained after treating at 400 ºC and 600 ºC, while after being treated at 800℃, the screen bent, and the back cover of iPhone melted. Increasing the combustion temperature to 1000℃, the screen scattered, and most of the components turned into ashes. Combustion enhanced the leachability of REEs, while the leachability of the other valuable metals, except for Zn, was barely affected. Most of the REEs present in the original iPhones occurred as oxidizable forms. With elevations in the combustion temperature up to 600 ºC, the oxidizable REEs were transformed to acid soluble forms. However, further elevations in temperature resulted in decreases in the acid soluble fraction and corresponding increases in the reducible and oxidizable forms. Additionally, combustion temperature also significantly altered the occurrence modes of other metals present in the iPhones. f) Synthetic MSWI Ash: It was found that in the absence of hydrogen peroxide, all the elements except for Si were leached to certain degrees. It is noteworthy that approximately 80% of Zn was leached with 1.2 M HCl. When hydrogen peroxide was added to the reaction system, noticeable increases in the leaching recovery of Fe, Mn, Co, Ni, and Cu were observed. The leaching recovery of Al and Si was barely affected by adding hydrogen peroxide. These results suggested that the majority of Zn in the synthetic MSWI ash existed as metal oxide, a portion of Fe, Mn, Co, Ni, and Cu existed as metal oxide, and Al and Si are associated with glasses which are difficult to leach. Additionally, the remaining Fe, Mn, Co, Ni, and Cu in the metallic form were efficiently oxidized in the presence of hydrogen peroxide. g) Pozzolanic Activity and Thermal Stability of MSWI Ash: MSWI fly ash has higher pozzolanic activity compared to the bottom ash sample, which indicates that the fly ash sample consumed more portlandite because of its smaller particle size as reactivity fundamentally relates to reaction surface area. However, after the recovery of valuable elements, the pozzolanic activity of both the valuable elements fraction and the less valuable elements-rich products decreased significantly, which means that the valuable elements recovery lowers the Ca(OH)2 consumption, thus leading to the low activity of SCM. h) Flowsheet Design for Metal Recovery from MSWI Ash: Based on the results of the comprehensive physical separation and acid leaching tests, circuits that enable the beneficiation of the valuable metals were developed. In these circuits, the valuable metals are recovered into nonferrous, ferrous, and other valuable metal concentrates, which are processed separately in the acid leaching step. The subsequent separation and purification steps are simplified due to the physical beneficiation step. In addition, the overall recovery cost is reduced since physical beneficiation is much cheaper compared with chemical processing. Using different technologies, such as selective precipitation and solvent extraction, a comprehensive hydrometallurgical circuit was designed, and compounds of Cu, Zn, Mn, Co, and Ni with a purity close to or even higher than 95% were successfully generated.

36 MATERIALS SCIENCE↗
NASA NTRS2015

Carbon and Oxygen Isotope Measurements of Ordinary Chondrite (OC) Meteorites from Antarctica Indicate Distinct Carbonate Species Using a Stepped Acid Extraction Procedure

The purpose of this study is to characterize the stable isotope values of terrestrial, secondary carbonate minerals from five Ordinary Chondrite (OC) meteorites collected in Antarctica. These samples were identified and requested from NASA based upon their size, alteration history, and collection proximity to known Martian meteorites. They are also assumed to be carbonate-free before falling to Earth. This research addresses two questions involving Mars carbonates: 1) characterize terrestrial, secondary carbonate isotope values to apply to Martian meteorites for isolating in-situ carbonates, and 2) increase understanding of carbonates formed in cold and arid environments with Antarctica as an analog for Mars. Two samples from each meteorite, each approximately 0.5 grams, were crushed and dissolved in pure phosphoric acid for 3 sequential reactions: a) R times 0 for 1 hour at 30 degrees Centigrade (fine calcite extraction), b) R times 1 for 18 hours at 30 degrees Centigrade (course calcite extraction), and c) R times 2 for 3 hours at 150 degrees Centigrade (siderite and/or magnesite extraction). CO (sub 2) was distilled by freezing with liquid nitrogen from each sample tube, then separated from organics and sulfides with a TRACE GC using a Restek HayeSep Q 80/100 6 foot 2 millimeter stainless column, and then analyzed on a Thermo MAT 253 Isotope Ratio Mass Spectrometer (IRMS) in Dual Inlet mode. This system was built at NASA/JSC over the past 3 years and proof-tested with known carbonate standards to develop procedures, assess yield, and quantify expected error bands. Two distinct species of carbonates are found: 1) calcite, and 2) non-calcite carbonate (future testing will attempt to differentiate siderite from magnesite). Preliminary results indicate the terrestrial carbonates are formed at approximately sigma (sup 13) C equal to plus 5 per mille, which is consistent with atmospheric CO (sub 2) sigma (sup 13) C equal to minus 7 per mille and fractionation of plus12 per mille based upon polar temperature of -20 degrees Centigrade. The oxygen values fractionate sigma (sup 18) O equal to minus 10-20 per mille lighter between the R times 0 and R times 1 reactions at 30 degrees Centigrade. The carbonate oxygen isotope measurements are consistently heavier than expected with meteoric water and temperatures from Antarctica, perhaps due to secondary carbonate formation during curation in Houston, TX.

Evans, Michael E.↗
DOE OSTI2022

A study of degradation mechanisms in PVDF-based photovoltaic backsheets

Abstract Commercial backsheets based on polyvinylidene fluoride (PVDF) can experience premature field failures in the form of outer layer cracking. This work seeks to provide a better understanding of the changes in material properties that lead to crack formation and find appropriate accelerated tests to replicate them. The PVDF-based backsheet outer layer can have a different structure and composition, and is often blended with a poly(methyl methacrylate) (PMMA) polymer. We observed depletion of PMMA upon aging with sequential (MAST) and combined (C-AST) accelerated stress testing. In field-aged samples from Arizona and India, where PVDF crystallizes in its predominant α-phase, the degree of crystallinity greatly increased. MAST and C-AST protocols were, to some extent, able to replicate the increase in crystallinity seen in PVDF after ~ 7 years in the field, but no single-stress test condition (UV, damp heat, thermal cycling) resulted in significant changes in the material properties. The MAST regimen used here was too extreme to produce realistic degradation, but the test was useful in discovering weaknesses of the particular PVDF-based outer layer structure studied. No excessive β-phase formation was observed after aging with any test condition; however, the presence of β-phase was identified locally by Fourier transform infrared spectroscopy (FTIR). We conclude that both MAST and C-AST are relevant tests for screening outdoor failure mechanisms in PVDF backsheets, as they were successful in producing material degradation that led to cracking.

14 SOLAR ENERGY↗
DOE OSTI2016

Tractable Experiment Design via Mathematical Surrogates

This presentation summarizes the development and implementation of quantitative design criteria motivated by targeted inference objectives for identifying new, potentially expensive computational or physical experiments. The first application is concerned with estimating features of quantities of interest arising from complex computational models, such as quantiles or failure probabilities. A sequential strategy is proposed for iterative refinement of the importance distributions used to efficiently sample the uncertain inputs to the computational model. In the second application, effective use of mathematical surrogates is investigated to help alleviate the analytical and numerical intractability often associated with Bayesian experiment design. This approach allows for the incorporation of prior information into the design process without the need for gross simplification of the design criterion. Illustrative examples of both design problems will be presented as an argument for the relevance of these research problems.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1980

As-built design specification for proportion estimate software subsystem

The Proportion Estimate Processor evaluates four estimation techniques in order to get an improved estimate of the proportion of a scene that is planted in a selected crop. The four techniques to be evaluated were provided by the techniques development section and are: (1) random sampling; (2) proportional allocation, relative count estimate; (3) proportional allocation, Bayesian estimate; and (4) sequential Bayesian allocation. The user is given two options for computation of the estimated mean square error. These are referred to as the cluster calculation option and the segment calculation option. The software for the Proportion Estimate Processor is operational on the IBM 3031 computer.

Obrien, S.↗
NASA NTRS1988

Radiation effects on polymeric materials

It is important to study changes in properties of polymeres after irradiation with charged particles, with ultraviolet radiation, and with combinations of both. An apparatus for this purpose has been built at the NASA Langley Research Center. It consists of a chamber 9 inches in diameter and 9 inches high with a port for an electron gun, another port for a mass spectrometer, and a quartz window through which an ultraviolet lamp can be focused. The chamber, including the electron gun and the mass spectrometer, can be evacuated to a pressure of 10 to the 8th power torr. A sample placed in the chamber can be irradiated with electrons and ultraviolet radiation separately, sequentially, or simultaneously, while volatile products can be monitored during all irradiations with the mass spectrometer. The apparatus described above has been used to study three different polymer films: lexan; a polycarbonate; P1700, a polysulfone; and mylar, a polyethylene terephthalate. All three polymers had been studied extensively with both electrons and ultraviolet radiation separately, but not simultaneously. Also, volatile products had not been monitored during irradiation for the materials. A high electron dose rate of 530 Mrads/hr was used so that a sufficient concentration of volatile products would be formed to yield a reasonable mass spectrum.

Kiefer, Richard L.↗
DOE OSTI2018

In situ monitoring of hydrogen loss during pyrolysis of wood by neutron imaging

Hydrogen is an element of fundamental importance for energy but hard to quantify in bulk materials. Neutron radiography was used to map in situ loss of elemental hydrogen from beech tree wood samples during pyrolysis. The samples consisted of three wood cylinders (finished dowel or cut branch) of approximately 1 cm in length. The samples were pyrolyzed under vacuum in a furnace vessel that was placed inside a cold neutron imaging beamline using a temperature ramp of 5 °C/min from ambient up to 400 °C. Neutron radiographs with exposures of 30 s were sequentially recorded with a charge-coupled device over the course of the experiment. Relative absorbance/scattering of the neutron beam by each sample was based on intensity (or brightness) values as a function of pixel position. The much larger neutron cross section for hydrogen compared to carbon and oxygen enables almost direct conversion of neutron attenuation into sample hydrogen content for each time step during the pyrolysis experiment. Target and vessel temperatures were recorded concurrently with collection of the radiographs so that changes could be directly correlated to different states of pyrolysis. The most visible change appeared at the initial phase of the 400 °C plateau as evidenced by strong hydrogen loss and primarily diametric shrinking of the samples. In conclusion, the loss of elemental hydrogen between initial and final states of pyrolysis was estimated to be about 70%.

08 HYDROGEN↗
DOE OSTI2018

Fabrication of Biodegradable Superabsorbent Using RSM Design for Controlled Release of KNO{sub 3}

Present study envisaged the sequential experimental design approach for the development of biodegradable Gelatin-Tapoica/polyacrylamide superabsorbent. Percentage water uptake efficacy of candidate sample was optimized using Response Surface Methodology (RSM) design under microwave irradiation. Different process variables such as potassium persulphate and ammonium persulphate (KPS:APS) ratio, pH, reaction time concentration of acrylamide and N,N-methylene-bis-acrylamide (MBA) were investigated as a function of percentage swelling using sequential experimental design. Maximum liquid efficacy of 1550% was obtained at KPS:APS = 1.0:0.5; acrylamide = 7.67 × 10{sup −1} mol L{sup −1}; MBA = 1.76 × 10{sup −2} mol L{sup −1}; pH 10 and time = 110 s. The 3D crosslinked network formed was characterized using Fourier Transformation Infrared spectroscopy (FTIR), X-ray diffraction (XRD), Scanning Electron Microscopic (SEM) techniques and thermal stability was ensured by Thermal gravimetric Analysis/Differential Thermal Analysis/Differential Thermal Gravimetric (TGA/DTA/DTG) studies. Superabsorbent synthesized could increase the moisture content in different type of soils and was found to enhance the water-holding capability of the soil upto 60 days in clayey, 40 days in sandy and 51 days in mixture of two soils under controlled conditions. Further, candidate polymer was investigated for the in-vitro controlled release of the KNO{sub 3} with diffusion exponent ‘n’ was found to be 0.4326 indicating Fickian type diffusion. Also, initial diffusion coefficient (D{sub I} = 3.49 × 10{sup −5} m{sup 2} h{sup −1}) was found to be greater than the lateral diffusion coefficient (D{sub L} = 3.76 × 10{sup −6} m{sup 2} h{sup −1}) indicated rapid release of KNO{sub 3} during initial hours with slow release afterwards. The ecofriendly nature of the synthesized polymer was also tested by conducting biodegradation studies and it was found to degrade upto 94% and 88.1% within 70 days with degradation rate of 1.34 and 1.26% per day using composting method and vermicomposting method respectively. So, the synthesized candidate polymer was found to be boon for agriculture-horticulture sector with wide applicability.

36 MATERIALS SCIENCE↗
DOE OSTI2014

Unraveling resistive versus collisional contributions to relativistic electron beam stopping power in cold-solid and in warm-dense plasmas

We present results on laser-driven relativistic electron beam propagation through aluminum samples, which are either solid and cold or compressed and heated by laser-induced shock. A full numerical description of fast electron generation and transport is found to reproduce the experimental absolute K{sub α} yield and spot size measurements for varying target thicknesses, and to sequentially quantify the collisional and resistive electron stopping powers. The results demonstrate that both stopping mechanisms are enhanced in compressed Al samples and are attributed to the increase in the medium density and resistivity, respectively. For the achieved time- and space-averaged electronic current density, 〈j{sub h}〉∼8×10{sup 10} A/cm{sup 2} in the samples, the collisional and resistive stopping powers in warm and compressed Al are estimated to be 1.5 keV/μm and 0.8 keV/μm, respectively. By contrast, for cold and solid Al, the corresponding estimated values are 1.1 keV/μm and 0.6 keV/μm. Prospective numerical simulations involving higher j{sub h} show that the resistive stopping power can reach the same level as the collisional one. In addition to the effects of compression, the effect of the transient behavior of the resistivity of Al during relativistic electron beam transport becomes progressively more dominant, and for a significantly high current density, j{sub h}∼10{sup 12} A/cm{sup 2}, cancels the difference in the electron resistive stopping power (or the total stopping power in units of areal density) between solid and compressed samples. Analytical calculations extend the analysis up to j{sub h}=10{sup 14} A/cm{sup 2} (representative of the full-scale fast ignition scenario of inertial confinement fusion), where a very rapid transition to the Spitzer resistivity regime saturates the resistive stopping power, averaged over the electron beam duration, to values of ∼1 keV/μm.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Lens Modeling of STRIDES Strongly Lensed Quasars Using Neural Posterior Estimation

Strongly lensed quasars can be used to constrain cosmological parameters through time-delay cosmography. Models of the lens masses are a necessary component of this analysis. To enable time-delay cosmography from a sample of $\mathcal{O}(10^3)$ lenses, which will soon become available from surveys like the Rubin Observatory’s Legacy Survey of Space and Time and the Euclid Wide Survey, we require fast and standardizable modeling techniques. To address this need, we apply neural posterior estimation (NPE) for modeling galaxy-scale strongly lensed quasars from the Strong Lensing Insights into the Dark Energy Survey (STRIDES) sample. NPE brings two advantages: speed and the ability to implicitly marginalize over nuisance parameters. We extend this method by employing sequential NPE to increase precision of mass model posteriors. We then fold individual lens models into a hierarchical Bayesian inference to recover the population distribution of lens mass parameters, accounting for out-of-distribution shift. After verifying our method using simulated analogs of the STRIDES lens sample, we apply our method to 14 Hubble Space Telescope single-filter observations. We find the population mean of the power-law elliptical mass distribution slope, γ lens , to be $\mathcal{M}_γ$ lens = 2.13 ± 0.06. Our result represents the first population-level constraint for these systems. This population-level inference from fully automated modeling is an important stepping stone toward cosmological inference with large samples of strongly lensed quasars.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

The Characterization of Dissolved Organic Matter in Reclaimed Water and Its Influence on Copper Toxicity

Water samples were collected from five phases of treatment in a municipal sewage reclaimed water plant and the DOM was characterized. Results indicated that the components and properties of DOM varied notably with sequential treatments, such as the fluorescence intensity, the molecular weight and the total acidity. Meanwhile the accumulation of Cu in Daphnia magna was analyzed following exposure to samples spiked with 50 µg/L copper, which were decreased by the presence of DOM in those water samples. Furthermore, this study found significant associations between fluorescence intensity and Cu accumulation (r{sup 2} = 0.778, p < 0.05), while increased total acidity was found to enhance the unit total organic carbon-Cu accumulation (r{sup 2} = 0.979, p < 0.01). This study provides useful information on the safety and effective management of reclaimed water as a potential water resource.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2014

A rigorous sequential update strategy for parallel kinetic Monte Carlo simulation

The kinetic Monte Carlo (kMC) method is used in many scientific fields in applications involving rare-event transitions. Due to its discrete stochastic nature, efforts to parallelize kMC approaches often produce unbalanced time evolutions requiring complex implementations to ensure correct statistics. In the context of parallel kMC, the sequential update technique has shown promise by generating high quality distributions with high relative efficiencies for short-range systems. In this work, we provide an extension of the sequential update method in a parallel context that rigorously obeys detailed balance, which guarantees exact equilibrium statistics for all parallelization settings. Furthermore, our approach also preserves nonequilibrium dynamics with minimal error for many parallelization settings, and can be used to achieve highly precise sampling.

36 MATERIALS SCIENCE↗
DOE OSTI2016

Fukushima Daiichi reactor source term attribution using cesium isotope ratios from contaminated environmental samples

Source term attribution of environmental contamination following the Fukushima Daiichi Nuclear Power Plant (FDNPP) disaster is complicated by a large number of possible similar emission source terms (e.g. FDNPP reactor cores 1–3 and spent fuel ponds 1–4). Cesium isotopic analyses can be utilized to discriminate between environmental contamination from different FDNPP source terms and, if samples are sufficiently temporally resolved, potentially provide insights into the extent of reactor core damage at a given time. In this work, rice, soil, mushroom, and soybean samples taken 100–250 km from the FDNPP site were dissolved using microwave digestion. Radiocesium was extracted and purified using two sequential ammonium molybdophosphate-polyacrylonitrile columns, following which 135 Cs/ 137 Cs isotope ratios were measured using thermal ionization mass spectrometry (TIMS). Results were compared with data reported previously from locations to the northwest of FDNPP and 30 km to the south of FDNPP. 135 Cs/ 137 Cs isotope ratios from samples 100–250 km to the southwest of the FDNPP site show a consistent value of 0.376 ± 0.008. 135 Cs/ 137 Cs versus 134 Cs/ 137 Cs correlation plots suggest that radiocesium to the southwest is derived from a mixture of FDNPP reactor cores 1, 2, and 3. Conclusions from the cesium isotopic data are in agreement with those derived independently based upon the event chronology combined with meteorological conditions at the time of the disaster. In conclusion, cesium isotopic analyses provide a powerful tool for source term discrimination of environmental radiocesium contamination at the FDNPP site. For higher precision source term attribution and forensic determination of the FDNPP core conditions based upon cesium, analyses of a larger number of samples from locations to the north and south of the FDNPP site (particularly time-resolved air filter samples) are needed. Published in 2016. This article is a U.S. Government work and is in the public domain in the USA.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2023

Deep reinforcement learning for class imbalance fault diagnosis of equipment in nuclear power plants

In equipment fault diagnosis in nuclear power plants, there may be far more samples in one class (e.g., a health state) than in another class (e.g., a fault state). The distribution of data in each class is highly skewed. Most machine learning algorithms are suitable for balanced training datasets. When faced with imbalanced samples, these algorithms tend to provide good identification for the majority classes and bias for the minority classes. However, the misclassification of minority classes can lead to high costs. To address the above problem, this paper develops a deep reinforcement learning-based diagnosis method that models fault diagnosis as a sequential decision-making process. At each time step, the agent receives the state of the environment represented by the training samples and then takes a diagnosis action guided by a policy. If the action is correct/incorrect, the agent receives a positive/negative reward. The reward for minority classes is higher than that for majority classes. The agent’s goal is to obtain as many cumulative rewards as possible in the process, i.e., to identify the sample as correctly as possible. Six demonstration scenarios are constructed, depending on the selected fault datasets and the designed model structures. Experiments show that the proposed method achieves a higher weighted-averaged F1 score than the classical supervised learning method in most cases of class imbalance. Finally, the proposed method has potential applications in the field of class imbalance fault diagnosis of equipment in nuclear power plants.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2018

Enhanced biofilm formation aids adaptation to extreme warming and environmental instability in the diatom Thalassiosira pseudonana and its associated bacteria

Abstract Recent studies have highlighted the capacity of marine diatoms at the foundation of ocean food webs and biogeochemical cycles to respond to environmental change. These responses are a combination of physiological acclimation and evolutionary responses. The latter occur rapidly within a few hundred generations (translating to a scale of months to years), due to the large standing genetic variation and short generation times of diatom populations. Studies are usually carried out on diatom species in their planktonic state, but this may constitute a stark oversimplification: diatoms readily form biofilms in the water column of shallow coastal areas, where biofilm formation affects carbon export, and in the benthos, where biofilms contribute significantly to sediment stability. Here, I have carried out a 400‐generation selection experiment, where I evolved the diatom Thalassiosira pseudonana and its associated bacteria. Bacteria and diatom were cultured in their planktonic form, and under strong selection for rapid biofilm formation in temperature and nutrient regimes differing in their severity and stability. I find that biofilm formation is enhanced under mild warming and rapid thermal fluctuations but not under extreme warming regardless of nutrient availability. Evolutionary responses are slower in biofilms under benign conditions but speed up under extreme and fluctuating conditions. Taxonomic composition of the associated bacteria is driven by the temperature selection regime in biofilms but not in planktonic samples. I also show that T. pseudonana , when selected for biofilm formation, presents predictable and sequential morphologies that are not usually observed in planktonic T. pseudonana.

Schaum, C. ‐E.↗