Search NASA⌕ Search

SEARCH · Search NASA

Results for “filter optimization”

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 1,099 records · Page 61

WE-AB-204-01: Performance Characterization of Regularized-Reconstruction Algorithm for 90Y PET/CT Images

Purpose: {sup 90}Y PET/CT imaging and quantification have recently been suggested as an approach of treatment verification. However, due to low positron yield (32ppm), the {sup 90}Y-PET/CT images are very noisy. Iterative reconstruction techniques that employ regularization, e.g. block sequential regularized expectation maximization (BSREM) algorithm (recently implemented on GE scanners – QClear™), has the potential to increase quantitative accuracy with lower noise penalty compared to OSEM. Our aim is to investigate the performance of RR algorithms in {sup 90}Y PET/CT studies. Methods: A NEMA IEC phantom filled with 3GBq {sup 90}YCl{sub 2} (to simulate patient treatment) was imaged on GE-D690 for 1800s/bed. The sphere-to-background ratio of 7. The data were reconstructed using OSEM and BSREM with PSF modeling and TOF correction while varying the iterations (IT) from 1–6 with fixed 24subsets. For BSREM, the edge-preservation parameter (γ ) was 2 and the penalty-parameters (β) was varied 350–950. In all cases a post-reconstruction filter of 5.2mm (2pixel) transaxial and standard z-axis were used. Sphere average activity concentration (AC) and background standard deviation (SD) were then calculated from VOIs drawn in the spheres and background. Results: Increasing IT from 1to6, the %SD in OSEM increased from 30% to 80%, whereas %SD in BSREM images increased by <5% for all βs. BSREM with β=350 didn’t offer any improvement over OSEM (convergence of mean achieved at 2 IT, in this study). Increasing β from 350 to 950 reduced the AC accuracy of small spheres (<20mm) by 10% and noise from 40% to 20%, which resulted in CNR increase from 11 to 17. Conclusion: In count-limited studies such as {sup 90}Y PET/CT, BSREM can be used to suppress image noise and increase CNR at the expense of a relatively small decrease of quantitative accuracy. The BSREM parameters need to be optimized for each study depending on the radionuclides and count densities. Research reported in this publication was supported by the National Cancer Institute of the National Institutes of Health under Award Number R01CA13898 and was also partly supported by General Electric.

60 APPLIED LIFE SCIENCES↗

TU-FG-BRB-09: Thermoacoustic Range Verification with Perfect Co-Registered Overlay of Bragg Peak onto Ultrasound Image

Purpose: The potential of particle therapy has not yet been fully realized due to inaccuracies in range verification. The purpose of this work was to correlate the Bragg peak location with target structure, by overlaying thermoacoustic localization of the Bragg peak onto an ultrasound image. Methods: Pulsed delivery of 50 MeV protons was accomplished by a fast chopper installed between the ion source and the inflector of the 88″ cyclotron at Lawrence Berkeley National Lab. 2 Gy were delivered in 2 µs by a beam with peak current of 2 µA. Thermoacoustic emissions were detected by a cardiac array and Verasonics V1 ultrasound system, which also generated a grayscale ultrasound image. 1024 thermoacoustic pulses were averaged before filtering and one-way beamforming focused signal onto the Bragg peak location with perfect co-registration to the ultrasound images. Data was collected in a room temperature water bath and gelatin phantom with a cavity designed to mimic the intestine, in which gas pockets can displace the Bragg peak. Experiments were performed with the cavity both empty and filled with olive oil. Results: In the waterbath overlays of the Bragg peak agreed with Monte Carlo simulations to within 800±170 µm. Agreement within 1.3 ± 0.2 mm was achieved in the gelatin phantom, although relative stopping powers were estimated only to first order from CT scans. Protoacoustic signals were detected after travel from the Bragg peak through 29 mm and 65 mm of phantom material when the cavity was empty and full of olive oil, respectively. Conclusion: Protoacoustic range verification is feasible with a commercial clinical ultrasound array, but at doses exceeding the clinical realm. Further optimization of both transducer array and injection line chopper is required to enable range verification within a 2 Gy dose limit, which would enable online adaptive treatment. This work was supported in part by a UWM Intramural Instrumentation Grant and by the Director, Office of Science, Office of Nuclear Physics, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231. YMQ was supported by a UWM-OUR summer fellowship.

60 APPLIED LIFE SCIENCES↗

Model-Based Sensor Placement for Component Condition Monitoring and Fault Diagnosis in Fossil Energy Systems

Design and operation of energy producing, near “zero-emission” coal plants has become a national imperative. This report on model-based sensor placement describes a transformative two-tier approach to identify the optimum placement, number, and type of sensors for condition monitoring and fault diagnosis in fossil energy system operations. The algorithms are tested on a high fidelity model of the integrated gasification combined cycle (IGCC) plant. For a condition monitoring network, whether equipment should be considered at a unit level or a systems level depends upon the criticality of the process equipment, its likeliness to fail, and the level of resolution desired for any specific failure. Because of the presence of a high fidelity model at the unit level, a sensor network can be designed to monitor the spatial profile of the states and estimate fault severity levels. In an IGCC plant, besides the gasifier, the sour water gas shift (WGS) reactor plays an important role. In view of this, condition monitoring of the sour WGS reactor is considered at the unit level, while a detailed plant-wide model of gasification island, including sour WGS reactor and the Selexol process, is considered for fault diagnosis at the system-level. Finally, the developed algorithms unify the two levels and identifies an optimal sensor network that maximizes the effectiveness of the overall system-level fault diagnosis and component-level condition monitoring. This work could have a major impact on the design and operation of future fossil energy plants, particularly at the grassroots level where the sensor network is yet to be identified. In addition, the same algorithms developed in this report can be further enhanced to be used in retrofits, where the objectives could be upgrade (addition of more sensors) and relocation of existing sensors.

20 FOSSIL-FUELED POWER PLANTS↗

TULIP: An RNA-seq-based Primary Tumor Type Prediction Tool Using Convolutional Neural Networks

Background: With cancer as one of the leading causes of death worldwide, accurate primary tumor type prediction is critical in identifying genetic factors that can inhibit or slow tumor progression. There have been efforts to categorize primary tumor types with gene expression data using machine learning, and more recently with deep learning, in the last several years. Methods In this paper, we developed four 1-dimensional (1D) Convolutional Neural Network (CNN) models to classify RNA-seq count data as one of 17 highly represented primary tumor types or 32 primary tumor types regardless of imbalanced representation. Additionally, we adapted the models to take as input either all Ensembl genes (60,483) or protein coding genes only (19,758). Unlike previous work, we avoided selection bias by not filtering genes based on expression values. RNA-seq count data expressed as FPKM-UQ of 9,025 and 10,940 samples from The Cancer Genome Atlas (TCGA) were downloaded from the Genomic Data Commons (GDC) corresponding to 17 and 32 primary tumor types respectively for training and validating the models. Results: All 4 1D-CNN models had an overall accuracy of 94.7% to 97.6% on the test dataset. Further evaluation indicates that the models with protein coding genes only as features performed with better accuracy compared to the models with all Ensembl genes for both 17 and 32 primary tumor types. For all models, the accuracy by primary tumor type was above 80% for most primary tumor types. Conclusions: We packaged all 4 models as a Python-based deep learning classification tool called TULIP (TUmor CLassIfication Predictor) for performing quality control on primary tumor samples and characterizing cancer samples of unknown tumor type. Further optimization of the models is needed to improve the accuracy of certain primary tumor types.

Jones, Sara↗

Selective capture and recovery of uranium oxide colloids from aqueous soil suspensions using high gradient magnetic filtration

High Gradient Magnetic Filtration (HGMF) is a promising method for the selective capture and recovery of uranium oxide from surface soils. To date, however, magnetic filtration of uranium oxide has only been demonstrated at a proof-of-principle scale using relatively small filters (<5 cm 3 ) at low flowrates (<60 mL/min). Here, to explore the efficacy of magnetic filtration of uranium oxide at a larger scale, a newly designed HGMF apparatus that is more than an order of magnitude larger than our earlier filters (106 cm 3 ) was designed, fabricated, and tested at relatively high flowrates. Filtration experiments were performed using aqueous uranium oxide particle suspensions with Arizona Road Dust (ARD) as a soil simulant. At a flowrate of 125 mL/min, the apparatus’ uranium capture rate was exceptionally high (96 %), but selectivity was poor due to the high rate of capture for diamagnetic soil constituents (e.g., 77 % for silicon). All particles were captured at a lower rate when the flowrate was increased to 250 mL/min, but uranium selectivity was significantly increased due to the more substantial reduction in diamagnetic particle capture (i.e., capture rate of 77 % and 15 % for uranium and silicon, respectively). When backwashing the apparatus at the same flowrates used during filtration experiments, the rate of uranium recovery tended to be fairly low. Nevertheless, higher flowrates (1 L/min) and sonication were both shown to be highly effective methods of increasing uranium recovery. Magnetic field simulations were also performed to investigate potential optimizations to the design of the apparatus. These simulations showed that the intensity of the applied magnetic field could be increased by increasing the thickness of the steel magnetic housing. Additionally, stochastic trajectory simulations were performed to investigate the potential mechanisms of particle capture.

HGMF↗

High-throughput assessment of hypothetical zeolite materials for their synthesizeability and industrial deployability

Zeolites are important microporous framework materials, where 200+ structures are known to exist and many millions so-called hypothetical materials can be computationally created. Here, we screen the “Deem” database of hypothetical zeolite structures to find experimentally feasible and industrially relevant materials. We use established and existing criteria and structure descriptors (lattice energy, local interatomic distances, TTT angles), and we develop new criteria which are based on 5-th neighbor distances to T-atoms, tetrahedral order parameters (or, tetrahedrality), and porosity and channel dimensionality. Our filter funnel for screening the most attractive zeolite materials that we construct consists of nine different types of criteria and a total of 53 subcriteria. The funnel reduces the pool of candidate materials from initially >300,000 to 70 and 33, respectively, depending on the channel dimensionality constraint applied (2- and 3-dimensional vs. only 3-dimensional channels). Here, we find that it is critically important to define longer range and more stringent criteria such as the new 5-th neighbor distances to T-atoms and the tetrahedrality descriptor in order to succeed in reducing the huge pool of candidates to a manageable number. Apart from four experimentally achieved structures (BEC, BOG, ISV, SSF), all other candidates are hypothetical frameworks, thus, representing most valuable targets for synthesis and application. Detailed analysis of the screening data allowed us to also propose an exciting future direction how such screening studies as ours could be improved and how framework generating algorithms could be competitively optimized.

36 MATERIALS SCIENCE↗

Presentation to JAEA Visitors, July 24, 2012 [Slides]

There are 7 Reasons for characterizing spent fuel nuclide concentrations: (1) Provide IAEA with the capability to independently verify the mass of plutonium at any site that has spent fuel; (2) Shipper/receiver difference; (3) Determination of the input accountability mass of an electrochemical (pyro-chemical) processing facility; (4) Continuity of knowledge at spent fuel storage site; (5) Optimal reloads through knowledge of true actinide content; (6) Burnup credit for fuel transport and storage; and (7) Provide confidence to the public that the shipment of spent fuel around the world is being undertaken in a rigorous manner, assuring that material is not diverted during shipment. Types of fuel of interest are: Fast Reactor Fuel, Normal burnup low enriched uranium (LEU), Mixed oxide (MOX), Low burnup LEU, and Research reactor. MCNPX calculates the slowing down and the energy spread for a given time. The principles are: (1) Neutrons slow down in LSDS; (2) Slow neutrons capture preferentially on a fission resonance; (3) Fission emits high-energy neutrons; (4) High-energy neutrons detected in threshold detector (that is not sensitive to slow neutrons) (e.g., 238 U, 232 Th); and (5) The detection time is characteristic of the slow neutron energy. Following each PSR pulse, the entire signal is digitized in 2 nanosecond intervals, and subsequently analyzed Analysis involves applying a digital filter to extract fission pulses in a background of initial oscillations and noise throughout. Some of the noise is not random. Depending on the threshold setting and filter parameters, both the normalization and the trend change. Original PNNL LSDS design required a 1-meter radius cylindrical LSDS. Much higher efficiency would allow us to use a (1.2m) LSDS (this is the LANL LSDS size), or a less intense source, or a combination of both. We still estimate 10 16 total neutrons from the source to obtain 2-3% precision in 239 Pu and 235 U assay.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Technology demonstration for reducing mercury emissions from small-scale gold refining facilities.

Gold that is brought from artisanal and small-scale gold mining areas to gold shops for processing and sale typically contains 5-40% mercury. The uncontrolled removal of the residual mercury in gold shops by using high-temperature evaporation can be a significant source of mercury emissions in urban areas where the shops are located. Emissions from gold shop hoods during a burn can exceed 1,000 mg/m{sup 3}. Because the saturation concentration of mercury vapor at operating temperatures at the hood exhaust is less than 100 mg/m{sup 3}, the dominant component of the exhaust is in the form of aerosol or liquid particles. The U.S. Environmental Protection Agency (EPA), with technical support from Argonne National Laboratory (Argonne), has completed a project to design and test a technology to remove the dominant aerosol component in the emissions from gold shops. The objective was to demonstrate a technology that could be manufactured at low cost and by using locally available materials and manufacturing capabilities. Six prototypes designed by Argonne were locally manufactured, installed, and tested in gold shops in Itaituba and Creporizao, Brazil. The initial prototype design incorporated a pebble bed as the media for collecting the mercury aerosols, and a mercury collection efficiency of over 90% was demonstrated. Though achieving high efficiencies, the initial prototype was determined to have practical disadvantages such as excessive weight, a somewhat complex construction, and high costs (>US$1,000). To further simplify the construction, operation, and associated costs, a second prototype design was developed in which the pebble bed was replaced with slotted steel baffle plates. The system was designed to have flexibility for installation in various hood configurations. The second prototype with the baffle plate design was installed and tested in several different hood/exhaust systems to determine the optimal installation configuration. The significance of coagulation and collection of the mercury aerosols in exhaust ducts, which is dependent on the hood and collector configuration, was also evaluated. Prototype demonstration tests verified the theoretical basis for mercury aerosol capture that can be used to optimize the baffle plate design, flow rates, and hood exhaust ducts and plenum to achieve 80% or higher removal efficiencies. Results indicated that installation configuration significantly influences a system's capture efficiency. Configurations that retained existing inlet ducts resulted in system efficiencies of more than 80%, whereas installation configurations without inlet ducts significantly reduced capture efficiency. As an alternative to increasing the volume of inlet ducts, the number of baffle plates in the system baffle assembly could be doubled to increase efficiency. Recommended installation and operation procedures were developed on the basis of these results. A water-based mercury capture system developed in Indonesia for installation in smaller shops was also tested and shown to be effective for certain applications. The cost of construction and installation of the baffle plate prototype was approximately US$400. These costs were reported as acceptable by local gold shop owners and government regulators, and were significantly lower than the cost of an alternate charcoal/copper mesh mercury filter available in the region, which costs about US$10,000. A sampling procedure that consists of a particle filter combined with a vapor analyzer was demonstrated as an effective procedure for analyzing both the aerosol and vapor components of the mercury concentrations. Two key findings for enhancing higher mercury collection were identified. First, the aerosol/vapor mercury emissions must be given sufficient time for the mercury particles to coagulate to a size that can be readily captured by the baffle plates. An interval of at least 6 seconds of transit time between the point of evaporation and contact with the slotted baffle plates is recommended. Some particles will also deposit in the exhaust ducts between the point of evaporation and the baffle plates. Second, the slots in the baffle plates create jets that force the mercury particles to impinge and adhere on downstream surfaces. The baffle plates should closely follow the designs developed for this system to be most effective.

54 ENVIRONMENTAL SCIENCES↗

Development and Implementation of an Automatic Continuous Online Monitoring and Control Platform for Polymerization Reactions to Sharply Boost Energy and Resource Efficiency in Polymer Manufacturing

The project goal was to create an energy saving paradigm shift in how polymers are manufactured in the 21st century. It used Automatic Continuous Online Monitoring of Polymerization reactions (ACOMP) integrated for the first time with automatic active control to create the innovative ‘ACOMP/Control Interface’, or ‘ACOMP/CI’. ACOMP/CI will begin the transformation from old, inefficient processes into highly evolved, energy and resource efficient ones. The ACOMP platform is broadly applicable to many types of reactions and processes throughout the vast polymer industry. The industry provides materials for sectors such as automotive, aerospace, oil recovery, agriculture, paints, resins, adhesives, pharmaceuticals and therapeutic proteins, optics, electronics, lightweight building materials, and many more. The U.S. chemical industry is one of the last major sectors in which the U.S. has top global stature. It consumes 24% of all U.S. manufacturing energy, produces over $800B of product annually, supports 25% of the U.S. GDP and employs over 6 million people. It is also a major source of GHG emissions. Polymers make up approximately 30% of this sector. It is estimated that annually 60 TBtu of energy could be saved and 3 million tons less of GHG emissions produced by optimizing production in the polyolefin manufacturing sector alone. The project scope included first time design and prototyping of an ACOMP/CI, creation of active reaction controllers, and demonstration of control capabilities on ideal, low concentration polymerization reactions. All these elements of the scope were met, including advances and findings not originally anticipated. Extensions to more complex reactions, beyond the reactor capabilities of the current project ACOMP/CI, such as polyolefins and other high pressure/high temperature reactions, are being proposed in Fall 2017 to CESMII, a DoE based NNMI. The initial proposal was for a three year funded project, but this was reduced to a two year project and budget due to funding constraints. Hence, some of the original plans, such as adaptation of the ACOMP/CI to more relevant industrial processes, such as emulsion and dispersion technologies, could not be carried out. A third year of funding was requested at the end of the project, but DoE did not have resources to grant this. The sub-contractor Fluence Analytics (previously Advanced Polymer Monitoring Technologies, Inc) designed, prototyped, and commissioned a working ACOMP/CI by June 2015. The reaction characteristics to be automatically controlled were i) conversion kinetics, ii) molecular weight, iii) copolymer composition, and iv) simultaneous molecular weight and composition. A two pronged control strategy was used. The Tulane/Fluence group took a basic principles approach that did not rely on kinetic models. The LSU group took a more complex, non-linear model-oriented approach involving complete kinetic descriptions of the reaction system Each of these approaches proved successful in their own way. By April 2016 fully automatic control of conversion and weight average molecular weight, Mw, trajectories was achieved using the Tulane/Fluence (TF) basic principles controller. Similar results were obtained by the LSU non-linear model controller by August 2016. The demonstration system was aqueous free radical polymerization of acrylamide, Am. The control variables were temperature and semi-batch feed to the reactor of Am monomer and initiator. A demonstration of active manual conversion control in an industrial process using high solids in inverse emulsion polymerization of Am was achieved. During Summer 2016 the TF controller was used in conjunction with a chain transfer agent, another control variable, to automatically produce multi-modal molecular weight distributions, MWD, in a single reactor. Industrially, multi-modal MWD are produced by mixing products made in separate reactors, requiring significant extra time, energy, and reactor resources. Recognizing the industrial potential a patent on automatic production of multi-modal polymers was filed, and DoE acknowledged. In Fall 2016 the TF team developed a basic principles controller for copolymer composition and demonstrated it on aqueous free radical copolymerization of comonomers Am and styrene sulfonate, SS. TF then fused the Mw and composition controllers to achieve simultaneous control of both Mw and copolymer composition trajectories. Numerous simultaneous trajectories were demonstrated, including a trimodal composition distribution with constant Mw. Meanwhile, the LSU group developed a Kalman filter to improve the results of their automatic Mw and conversion controller and successful tests were made. During the project the TF team developed a means of computing full MWD during polymer synthesis without need for any chromatographic separation, based on model distributions. This means the polymer product is ‘born characterized’ and this can eliminate post-manufacture analytical laboratory quality control. TF filed a joint patent application on this new approach to chromatography-free determination of MWD with acknowledgment to DoE. The Tulane group obtained a 60MHz NMR during the project and recently completed the first work on separating three comonomers, Am, SS, and Na-acrylate, with a first demonstration of terpolymer composition control with the TF basic principles controller. Widespread dissemination of ACOMP/CI in the polymer manufacturing sector will bolster DoE goals of energy efficiency and reduced GHG emissions: The ability to monitor and actively control polymerization reactions will lead to more efficient use of energy and non-renewable resources, plant and labor time, increase the safety of manufacturing personnel, and will enhance product quality and lead to feasibility of manufacturing of polymers currently too complex for industrial scale production, while leading to less GHG emissions per kilo of product, and allowing for increased U.S. competitiveness in this enormous manufacturing sector. When ACOMP/CI is expanded to the polyolefin industry it is estimated that 60 TeraBTU/year of energy can be saved. Much of this saving is anticipated to come from optimized control of grade changeovers in steady state reactors and maintenance of steady states. Conclusions: ACOMP’s ability to provide continuous realtime data streams of measured polymer and reaction characteristics made it possible, for the first time, to directly and automatically control free radical polymerization reactions. An industrial client of Fluence Analytics has requisitioned the first ACOMP/CI which uses the TF basic principles controller. This sets the stage for FA to add control features to the ACOMP systems it has begun to install on the industrial scale beginning in 2014. Recommendations: This successful project was mainly limited to ideal polymerizations not of an industrial sort. The most energy intensive portion of polymer manufacturing is polyolefins. Adoption of ACOMP/CI to this enormous industrial sector faces the enormous challenges of high temperature, high pressure continuous sampling and high temperature sensor operation to obtain the continuous data needed for direct reaction control. The project team has a strategy for achieving this ambitious goal and will present it in Fall 2017 as a proposal to CESMII/DoE. It is recommended that this upcoming proposal be funded in order to make full use of the achievements of this just ended DoE project as the next step towards making polyolefin ACOMP/CI an energy saving reality. It is projected that ACOMP/CI can have its first polyolefin testbed demonstrations within two years of beginning the proposed project.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

AlphaBuilding ResCommunity: A multi-agent virtual testbed for community-level load coordination

Training and validating algorithms in a simulation testbed can accelerate research and applications of optimal control of residential loads to improve energy flexibility and grid resilience. We developed an open-source simulation environment, AlphaBuilding ResCommunity, that can be used to train and validate algorithms to control a single thermostatically controlled load (TCL) or coordinate a group of TCLs. We used reduced-order models to simulate the thermodynamics of TCLs, and the parameter values were determined from the connected smart thermostat data of real households. The environment was built upon the standardized OpenAI Gym interface. Ancillary functions, such as retrieving the parameters and weather forecasts, are provided to facilitate control strategies that require predictive information. Compared with existing efforts, AlphaBuilding ResCommunity has three advantages: (1) more realistic model settings because the parameter values are identified from actual household operating data, and modelling and measurement uncertainty are considered; (2) passive thermal storage control; and (3) ease of use due to a simple software dependency and standardized interface. We demonstrated the applications of the environment by implementing a Kalman Filter and Model Predictive Control on a single TCL and a Priority-Stack-Based Control and Alternating Direction Method of Multipliers to coordinate multiple TCLs for load tracking.

Wang, Z↗

WE-G-204-01: BEST IN PHYSICS (IMAGING): Effect of Image Processing Parameters On Nodule Detectability in Chest Radiography

Purpose: To investigate the effect of varying system image processing parameters on lung nodule detectability in digital radiography. Methods: An anthropomorphic chest phantom was imaged in the posterior-anterior position using a GE Discovery XR656 digital radiography system. To simulate lung nodules, a polystyrene board with 6.35mm diameter PMMA spheres was placed adjacent to the phantom (into the x-ray path). Due to magnification, the projected simulated nodules had a diameter in the radiographs of approximately 7.5 mm. The images were processed using one of GE’s default chest settings (Factory3) and reprocessed by varying the “Edge” and “Tissue Contrast” processing parameters, which were the two user-configurable parameters for a single edge and contrast enhancement algorithm. For each parameter setting, the nodule signals were calculated by subtracting the chest-only image from the image with simulated nodules. Twenty nodule signals were averaged, Gaussian filtered, and radially averaged in order to generate an approximately noiseless signal. For each processing parameter setting, this noise-free signal and 180 background samples from across the lung were used to estimate ideal observer performance in a signal-known-exactly detection task. Performance was estimated using a channelized Hotelling observer with 10 Laguerre-Gauss channel functions. Results: The “Edge” and “Tissue Contrast” parameters each had an effect on the detectability as calculated by the model observer. The CHO-estimated signal detectability ranged from 2.36 to 2.93 and was highest for “Edge” = 4 and “Tissue Contrast” = −0.15. In general, detectability tended to decrease as “Edge” was increased and as “Tissue Contrast” was increased. A human observer study should be performed to validate the relation to human detection performance. Conclusion: Image processing parameters can affect lung nodule detection performance in radiography. While validation with a human observer study is needed, model observer detectability for common tasks could provide a means for optimizing image processing parameters.

60 APPLIED LIFE SCIENCES↗

Sodium silicate treatment for uranium (vi) bearing groundwater systems at f/h area at savannah river site - 15709

The Savannah River Site (SRS) was one of the most significant manufacturing facilities during the Cold War era for producing nuclear materials. At the end of the Cold War, the Site's mission changed to support the environmental restoration of the Site due to over six decades of research, development, and production of nuclear weapons. Currently SRS is a major hazardous waste management facility responsible for nuclear materials storage, and remediation of contaminated soil and groundwater from radionuclides. This research focuses on controlling the mobilization of the contaminants, specifically uranium (VI) located in groundwater plumes at the Site's F/H Area Seepage Basin, where approximately 1.8 billion gallons of hazardous waste were deposited. Objective: To evaluate the potential use of sodium silicate for uranium removal from the aqueous phase, as well as to restore the pH of the treatment zone. Adding sodium silicate increases the pH of the treatment zone and uranium precipitation is achieved, therefore immobilizing the contaminant. Through a series of experiments the optimal concentration of sodium silicate was investigated. Part 1: An experiment was conducted in order to determine the effect of sodium silicate on the pH of a solution containing Savannah River Site soil and synthetic groundwater replicated after the Site's natural groundwater composition. Each sample was spiked with the appropriate volume from the sodium silicate stock solution in order to achieve final desired sodium silicate concentrations of 10, 20, 40,50,60,70, and 80 parts per million. The samples were placed on a platform shaker for three days during which pH readings were collected in regular intervals to determine which silicate concentration produced a pH in the neutral range (6 > pH < 8). Part 2: A sequential experiment was conducted specifically using 40 and 50 ppm of sodium silicate with Savannah River Site soil, synthetic groundwater, and 0.5 ppm of uranium (VI), in order to investigate the removal of uranium in the aqueous phase via precipitation. Samples were extracted and categorized into a filtered and non-filtered control group. Each sample was diluted 1:20 with 1% nitric acid and analyzed using the Kinetic Phosphorescence Analyzer (KPA) in order to measure uranium concentrations in the supernatant. By measuring the concentration of uranium in the supernatant the percent removal is obtained for the comparison between the filtered and non-filtered samples, revealing the amount of uranium precipitation that occurred for each silicate concentration. Results: A sodium silicate concentration of between 40-50 ppm proved ideal for achieving uranium precipitation and creating a circum-neutral pH environment in the aqueous phase. 50 ppm achieved the larger percent of uranium precipitation with a range of 30-50% removal for the filtered samples and between 60-95% removal for the non-filtered samples. A large difference between the removal of filtered and non- filtered samples from the same day was observed for both sodium silicate concentrations tested, indicating that there is a considerable amount of uranium species retained in the filter. The addition of sodium silicate proved effective in the removal of uranium precipitation from the aqueous phase. Future work: Determine potential re-solubilization of uranium concentrations; Examine mineralogy and surface morphology of dried precipitate using XRD and SEM/EDS.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

A State-Space Model of an Inverter-Based Microgrid for Multivariable Feedback Control Analysis and Design

In this work, a synchronous model for grid-connected and islanded microgrids is presented. The grid-connected model is based on the premise that the reference frame is synchronized with the AC bus. The quadrature component of the AC bus voltage can be cancelled, which allows to express output power as a linear equation for nominal values in the AC bus amplitude voltage. The model for the islanded microgrid is developed by integrating all the inverter dynamics using a state-space model for the load currents. This model is presented in a comprehensive way such that it could be scalable to any number of inverter-based generators using inductor–capacitor–inductor (LCL) output filters. The use of these models allows designers to assess microgrid stability and robustness using modern control methods such as eigenvalue analysis and singular value diagrams. Both models were tested and validated in an experimental setup to demonstrate their accuracy in describing microgrid dynamics. In addition, three scenarios are presented: non-controlled model, Linear-Quadratic Integrator (LQI) power control, and Power-Voltage (PQ/Vdq) droop–boost controller. Experimental results demonstrate the effectiveness of the control strategies and the accuracy of the models to describe microgrid dynamics.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Characterization of masses in digital breast tomosynthesis: Comparison of machine learning in projection views and reconstructed slices

Purpose: In digital breast tomosynthesis (DBT), quasi-three-dimensional (3D) structural information is reconstructed from a small number of 2D projection view (PV) mammograms acquired over a limited angular range. The authors developed preliminary computer-aided diagnosis (CADx) methods for classification of malignant and benign masses and compared the effectiveness of analyzing lesion characteristics in the reconstructed DBT slices and in the PVs. Methods: A data set of MLO view DBT of 99 patients containing 107 masses (56 malignant and 51 benign) was collected at the Massachusetts General Hospital with IRB approval. The DBTs were obtained with a GE prototype system which acquired 11 PVs over a 50 deg. arc. The authors reconstructed the DBTs at 1 mm slice interval using a simultaneous algebraic reconstruction technique. The region of interest (ROI) containing the mass was marked by a radiologist in the DBT volume and the corresponding ROIs on the PVs were derived based on the imaging geometry. The subsequent processes were fully automated. For classification of masses using the DBT-slice approach, the mass on each slice was segmented by an active contour model initialized with adaptive k-means clustering. A spiculation likelihood map was generated by analysis of the gradient directions around the mass margin and spiculation features were extracted from the map. The rubber band straightening transform (RBST) was applied to a band of pixels around the segmented mass boundary. The RBST image was enhanced by Sobel filtering in the horizontal and vertical directions, from which run-length statistics texture features were extracted. Morphological features including those from the normalized radial length were designed to describe the mass shape. A feature space composed of the spiculation features, texture features, and morphological features extracted from the central slice alone and seven feature spaces obtained by averaging the corresponding features from three to 19 slices centered at the central slice were compared. For classification of masses using the PV approach, a feature extraction process similar to that described above for the DBT approach was performed on the ROIs from the individual PVs. Six feature spaces obtained from the central PV alone and by averaging the corresponding features from three to 11 PVs were formed. In each feature space for either the DBT-slice or the PV approach, a linear discriminant analysis classifier with stepwise feature selection was trained and tested using a two-loop leave-one-case-out resampling procedure. Simplex optimization was used to guide feature selection automatically within the training set in each leave-one-case-out cycle. The performance of the classifiers was evaluated by the area (A{sub z}) under the receiver operating characteristic curve. Results: The test A{sub z} values from the DBT-slice approach ranged from 0.87{+-}0.03 to 0.93{+-}0.02, while those from the PV approach ranged from 0.78{+-}0.04 to 0.84{+-}0.04. The highest test A{sub z} of 0.93{+-}0.02 from the nine-DBT-slice feature space was significantly (p=0.006) better than the highest test A{sub z} of 0.84{+-}0.04 from the nine-PV feature space. Conclusion: The features of breast lesions extracted from the DBT slices consistently provided higher classification accuracy than those extracted from the PV images.

60 APPLIED LIFE SCIENCES↗

Scaling comb-driven resonator-based DWDM silicon photonic links to multi-Tb/s in the multi-FSR regime

The use of chip-based micro-resonator Kerr frequency combs in conjunction with dense wavelength-division multiplexing (DWDM) enables massively parallel intensity-modulated direct-detection data transmission with low energy consumption. Resonator-based modulators and filters used in such systems can limit the number of usable wavelength channels due to practical constraints on the maximum achievable free spectral range (FSR). In this work, we introduce the design of multi-Tb/s comb-driven resonator-based silicon photonic links by leveraging the multi-FSR regime. We demonstrate the viability of the link architecture with yield estimates that are supported by extensive wafer-scale measurements of 704 micro-resonators fabricated in a commercial complementary metal–oxide–semiconductor foundry. We show that a 2.80 Tb/s link is realizable with a ≥6 σ yield (∼99.999%), and that aggregate bandwidths of 3.76 Tb/s and 4.72 Tb/s are possible if yield targets are relaxed (3 σ and 1 σ , respectively). All designs represent a 1.94−3.28× boost to aggregate link bandwidth while maintaining BER≤10 −10 performance, with a theoretical bandwidth of 10.51 Tb/s being possible for sufficiently robust resonators. We use high-speed BER measurements to inform co-optimization of data rate and aggressor spacing ( λ ag ), limiting any additional loss-based power penalties to off-resonance insertion loss (IL) and routing loss. This work demonstrates that, through the multi-FSR regime, there is a clear path toward Kerr comb-driven ultra-broadband, high bandwidth silicon photonic links that can support next-generation data centers and high-performance computers.

James, Aneek (ORCID:0000000262527807)↗

TU-F-BRE-01: A High Resolution Micro Fiber Scintillator Detector Optimized for SRS and SBRT in Vivo Real Time Treatment Verification

Purpose: We have built a high resolution real time scintillating fiber detector prototype to determine in real time the accuracy of stereotactic radiosurgery (SRS) and stereotactic body radiotherapy (SBRT) treatments when only a fraction of the planned dose was delivered. The motivation of this work is to enhance dose delivery accuracy and to achieve error free radiosurgery. Methods: A high density array of scintillating fibers and a high speed photo detectors array were integrated to implement a high resolution real time dosimeter that can sample with high resolution pulsed SRS and SBRT beams cross sections. The high efficiency of the developed system allows to read each linac pulse in real time and to compute the accumulated dose and dose errors when only a fraction of the beam was delivered. The fibers are highly packed in a substrate that is directly coupled to two 128 pixel arrays with a pitch matching the fiber spacing to achieve accurate spatial localization. The small cross section of the fiber array allows stacking multiple fiber arrays to measure independent angular profiles that are digitally processed in parallel for real time dosimetry. Results: We implemented a high density array detector prototype with a pitch of 0.5 mm, readout speed of 1.2 msec, and a response time of 0.5 usec. The fast reading speed has the capability to determining the dose in flattening free filter beams. The detector can be installed in transmission mode at the output port of a micro-MLC. Treatment deviations smaller than 3% are detected when less than 1/100 of the planned dose was delivered. Conclusions: We built a prototype of a high resolution fiber scintillator array detector for SRS and SBRT in vivo dosimetry. Results show that the developed detector has the potential to assure error free SRS and SBRT treatments.

60 APPLIED LIFE SCIENCES↗

Installing Deep Groundwater Monitoring Wells in a Volcanic Dominated Rift Basin: Los Alamos National Laboratory, New Mexico - 17594

Los Alamos National Laboratory (LANL, or the Laboratory) is located in northern New Mexico near the west margin of the Espanola Basin. LANL is underlain by a complex basin-fill sequence of alluvial fan deposits, volcanic tuff, basaltic and dacitic lavas, and riverine deposits. On-going environmental remediation and characterization activities have evolved to improve the performance of groundwater monitoring wells, ensuring they yield representative groundwater and core samples. Groundwater occurs as canyon-fill alluvial water, perched intermediate-depth water associated with the basalts and other local-scale perching horizons, and as a regional-scale aquifer that is the main water supply for the area. Anthropogenic contaminants are found in all three groundwater zones. The regional aquifer below LANL is nominally 1000-1400 feet below ground surface (bgs). Monitoring wells at the Laboratory, are therefore, among the deepest routinely installed in the US. Logistical challenges at LANL include steep terrain, widespread cultural sites, and threatened and endangered species habitat restrictions. Additionally, the cost of installing monitoring wells is high because of groundwater depths and the variable drilling techniques that are required to characterize and install wells in a complex geologic environment. A major challenge to the characterization approach at LANL has been to meet sample quality requirements while minimizing drilling costs. In the late 1990's threaded drill-casing was advanced by rotary drilling methods using compressed air as the only circulation fluid. This resulted in relatively low impacts to the geochemical integrity of rock and groundwater samples, but led to frequent episodes of stuck casing. Additional fluids were introduced to the air-rotary program, improving penetration rates, but also resulting in non-representative groundwater samples due to long-term impacts to the aquifer in the vicinity of the well screen. Mud-rotary drilling was incorporated into the program, but it also had significant long-term impacts on groundwater quality, and often inhibited identification of perched groundwater zones and accurate identification of the water table in the regional aquifer. Current drilling practice makes limited use of fluid additives (e.g. foam) to supplement air circulation methods in order to advance drill casing through the vadose zone. Starting about 100 feet above the water table, drill casing is advanced to the target horizon in the regional aquifer using only air and municipal water (when needed) for circulation. Additionally, the threaded drill casing couplings have been replaced with welded connections, significantly reducing the occurrence of stuck casing. Careful observations of water production and water levels during drilling, coupled with the ability to retract casing for video and geophysical logs, now allows for robust characterization of perched groundwater systems and accurate definition of the top of regional saturation. Recent advances in the drilling program include combined dual-rotary/sonic drilling to depths up to 1150 feet and dual-rotary casing advance at angles up to 25 deg. from vertical. The former allows for core collection with minimal fluid addition, thereby preserving pore water chemistry. The latter method allows for well installation into portions of the aquifer otherwise inaccessible due to constraints at the surface (e.g., cultural sites or topography). Similar to the development of drilling techniques, both well design and groundwater sampling system design have evolved throughout the program. Current well design emphasizes a minimal annulus of filter pack, maximum screen slot size based on formation sieve analyses, and thorough well development to mitigate formation damage due to drilling. To sample groundwater at multiple levels in the same well, earlier sampling systems included Westbay{sup R} multi-zone systems and dual-screen single-pump and dual-screen dual-pump systems designed in collaboration with Baski, Inc. These multi-zone sampling systems proved difficult to develop and maintain given the depth to groundwater, and most current well installations are configured with a single screen and a submersible pump. The ability to purge from a well-developed single-screen well has substantially improved the collection of representative groundwater samples. A carefully crafted groundwater characterization program in target aquifers at depths ranging from 1,000 to 1,400 feet bgs within a complex geologic setting, while also retaining the ability to collect representative core and groundwater samples, may have application at a number of sites throughout the environmental industry. Optimizing each well to meet program objectives while reducing total project costs benefits all environmental investigations. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Defect filtering for thermal expansion induced dislocations in III–V lasers on silicon

Epitaxially integrated III–V semiconductor lasers for silicon photonics have the potential to dramatically transform information networks, but currently, dislocations limit performance and reliability even in defect-tolerant InAs quantum dot (QD)-based lasers. Despite being below the critical thickness, QD layers in these devices contain previously unexplained misfit dislocations, which facilitate non-radiative recombination. We demonstrate here that these misfit dislocations form during post-growth cooldown due to the combined effects of (1) thermal-expansion mismatch between the III–V layers and silicon and (2) mechanical hardening in the active region. By incorporating an additional sub-critical thickness, indium-alloyed “misfit dislocation trapping layer,” we leverage these mechanical hardening effects to our advantage, displacing 95% of misfit dislocations from the QD layer in model structures. Unlike conventional dislocation mitigation strategies, the trapping layer reduces neither the number of threading dislocations nor the number of misfit dislocations. It simply shifts the position of misfit dislocations away from the QD layer, reducing the defects' impact on luminescence. In full lasers, adding a misfit dislocation trapping layer both above and below the QD active region displaces misfit dislocations and substantially improves performance: we measure a twofold reduction in lasing threshold currents and a greater than threefold increase in output power. Overall, our results suggest that devices employing both traditional threading dislocation reduction techniques and optimized misfit dislocation trapping layers may finally lead to fully integrated, commercially viable silicon-based photonic integrated circuits.

36 MATERIALS SCIENCE↗