Search NASA⌕ Search

SEARCH · Search NASA

Results for “consistent filtering”

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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

Sub-drain Water Treatment Facility for Reduction of Accumulated Waste Water Amount at the Fukushima NPS Site - 17572

At the Fukushima Daiichi Nuclear Power Station site, water treatment facility special for sub-drain and groundwater drain has been constructed by Tokyo Electric Power Company Holdings, Inc. and Hitachi-GE Nuclear Energy, Ltd., and it has been operated to remove radionuclides from the pumped groundwater. The facility consists of three kinds of equipment for collecting, purifying and transferring. The purification equipment removes radionuclides contained in the pumped groundwater. The purification equipment consists of filters for removal of solids and colloidal particles, and adsorbents for removal of ionic radionuclides. Five adsorption vessels are installed into the equipment; three vessels have adsorbent to remove Cs and Sr simultaneously, the fourth vessel has adsorbent to remove Sb and for the last has adsorbent to remove heavy metals. The vessels for Cs and Sr are operating as a so-called 'merry-go-round'. For the rational operation of the equipment, a simulation code has been developed to calculate the Sr concentration of outlet water which determines the replacement timings of the Cs-Sr adsorption vessels. The simulation code has been applied to planning of vessel replacement during the water treatment facility operation. The facility started operation in September 2015, and treated the pumped groundwater since then. As of October 2016, about 211,000 m{sup 3} of groundwater had been treated. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2016

RO concentrated water treatment equipment for risk reduction of contaminated water stored in tank in Fukushima NPS - 16655

At the Fukushima Daiichi Nuclear Power Station (NPS) site, concentrated waste water was stored in tanks, awaiting treatment to remove radionuclides using various multi-nuclide removal systems. To accelerate the treatment for risk reductions of radiation dose and tank leakage, we constructed reverse-osmosis (RO) concentrated water treatment equipment. The equipment consists of filters and five adsorbent vessels containing Cs-Sr simultaneous adsorbents with the removal ability of Sr being 1/1000(max) to 1/100(min), and the overall treatment capacity of 500 to 900 m{sup 3}/day. We developed a simulation code to calculate the Sr concentration of outlet water for each vessel in order to evaluate the bed life of the media. The simulation code makes it possible to plan when to change media in the vessels during the equipment operation. The successful treatment of concentrated waste water was conducted using the equipment from January to May 2015. Total amount of water treated was 82,194 m{sup 3}. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2013

A monolithic polarization-independent frequency-filter system for filtering of photon pairs

We set up a long-term stable filtering system that consists of cascaded monolithic Fabry-Pérot filters to enhance the suppression and free spectral range. An effective free spectral range of hundreds of GHz allows the system to be used as a high resolution monochromator, with a linewidth of 192 MHz. As an important application a single mode is filtered from photon pairs generated by a parametric down-conversion source, and their indistinguishability is proven by measuring the Hong-Ou-Mandel effect with a visibility of 96%. We report that undesired birefringence, which is often encountered with monolithic cavities, can be avoided by stress-free mounting.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2025

Structural, chemical, and electronic control in Co–SiNx granular metals for high-pass filter applications

Granular metals, consisting of nanoscale conducting and insulating regions, have been studied for more than 50 years for fundamental and applied research. Granular metals exhibit non-linear conductivity vs frequency behavior, consistent with the universal power law response, and have recently been suggested for high-pass filter applications. Here, we report that cobalt–silicon nitride (Co–SiNx) granular metals with optimized sputter conditions and post-growth annealing exhibit an exceptional 109 increase in conductivity at 1 MHz compared to the DC conductivity. The improved frequency response is correlated with structural and chemical improvements examined via scanning transmission electron microscopy and x-ray photoemission spectroscopy. While we focus on improvements for high-pass filter applications, the structural, chemical, and electronic control demonstrated here will benefit a variety of granular metal and nanoparticle applications.

Annealing↗
DOE OSTI2012

A Preliminary Analysis of Dose Rates Associated with ITER CVCS Equipment/Area Location

A preliminary analysis of the ITER Chemical and Volume Control System (CVCS) Area was performed to assess dose rates outside the walls and ceiling of the facility after 1.5 years of operation at shutdown, 2 days, and 10 days after shutdown. For this purpose a simplified Monte Carlo computer model was developed using the MCNP (MCNP5 Ver. 1.51) code. Two components are included: the smaller filter tank and the larger ion exchanger. These pieces of equipment are associated with the Integrated Blanket ELM Divertor Primary Heat Transfer System, which will have the largest dose rates associated with activated corrosion products during operation in comparison with other systems. The ion exchanger contained two source regions, a 1.2-m-thick resin bed above a 0.55 m-thick skirt, and a 0.8-m-thick water region. The filter constituted an additional source. Thus the model consisted of three sources (filter, resin, water), homogeneously distributed within the appropriate source regions. However, much of the results (that address individual isotopes) are presented with the two sources in the ion exchanger combined. In these cases the sources are referred to as the 'ion exchanger source' and the 'filter source.' Dimensions for the facility and components, as well as source isotopes and strengths, and material densities, were supplied by US ITER. Because of its simplification, the model does not contain pipes. Consequently, radiation streaming through pipe penetrations, radiation emanating from the pipes, and shielding from the pipes were not considered in this analysis. Dose rates on the outside of two walls and the ceiling were calculated. The two walls are labeled as the 'long' wall (aligned with the X-axis) and the 'short' wall (aligned with the Y-axis). These walls and ceiling were nominally set to 30-cm-thick concrete. In the original analysis, standard concrete (2.3 g/cc density) was used. In addition to the shielding walls/ceiling, a floor and an additional wall opposite the long wall were added for photon scattering contributions. These were both 10-cm-thick, standard concrete structures. Other components (tanks, pipes, etc.), that were not included in the model, would potentially add additional scattering and shielding. Possibly these additional effects will be addressed in a later, more detailed analysis. The room was 29.6 m in length (X-axis), but was limited to 15 m in the model. The inside width (Y axis) and height (Z axis) were 4 m and 3.4 m, respectively. The origin for the model was located inside the room at the corner opposite the long wall and adjacent to the short wall at the floor level. The room was filled with air at standard temperature and pressure. The stainless steel (SS304) wall thicknesses for the ion exchanger and filter were 2.2 cm and 0.8 cm, respectively. The axial center of the filter was located 140 cm from the short wall and 100 cm from the long wall (outer surface). The axial center of the ion exchanger was located 440 cm from the short wall and 250 cm from the long wall (inner surface). The resin was assumed to be a homogeneous mixture of equal atom density fractions of hydrogen and carbon* at a specified density of 1.136 g/cc. The filter material was assumed to be homogeneous carbon at a specified density of 1.8 g/cc. If the filter media were stainless steel and the accumulated activity were the same, the dose rate outside the filter would be lower, provided the density of the stainless steel sintered filter material is significantly higher than 1.8g/cc. The densities of the water and air were assumed to be 1.0 g/cc and 1.096E-3 g/cc, respectively. The model included 10 small volumes placed outside the model adjacent to the wall at locations where the dose rates were expected to be highest and which showed to a degree the fall-off of the dose rate with distance along a particular wall or ceiling. These 'tally cells' are shown in Fig. 2 and are described in Table 1. Each cell had dimensions of 50 cm x 50 cm x 1 cm and was oriented so that the 1-cm-thick dimension was perpendicular to the wall against which the cell was located. Also, smaller tally volumes were placed at 5 cm (updated model), 30 cm, 50 cm, and 100 cm from each source, both radially from the outside of each component at the height of each source mid-plane, and vertically, along the axial centerline, above the ion-exchanger and below the filter. (For the ion exchanger, the vertical distance was limited to 75 cm because of interference from the ceiling.) These tally regions can be seen in Fig. 3. Each cell had dimensions of 5 cm x 5 cm x 1 cm, oriented so that the 1-cm-thick dimension was perpendicular to the X, Y or Z axis.

08 HYDROGEN↗
DOE OSTI2012

Advanced Techniques for Removal of Retrievable Inferior Vena Cava Filters

Inferior vena cava (IVC) filters have proven valuable for the prevention of primary or recurrent pulmonary embolism in selected patients with or at high risk for venous thromboembolic disease. Their use has become commonplace, and the numbers implanted increase annually. During the last 3 years, in the United States, the percentage of annually placed optional filters, i.e., filters than can remain as permanent filters or potentially be retrieved, has consistently exceeded that of permanent filters. In parallel, the complications of long- or short-term filtration have become increasingly evident to physicians, regulatory agencies, and the public. Most filter removals are uneventful, with a high degree of success. When routine filter-retrieval techniques prove unsuccessful, progressively more advanced tools and skill sets must be used to enhance filter-retrieval success. These techniques should be used with caution to avoid damage to the filter or cava during IVC retrieval. This review describes the complex techniques for filter retrieval, including use of additional snares, guidewires, angioplasty balloons, and mechanical and thermal approaches as well as illustrates their specific application.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2024

Minimum entropy filtering for a single output non-Gaussian stochastic system using state transformation

This paper presents a novel filter design for the single-output stochastic non-linear systems subjected to non-Gaussian noises and the proposed assumptions. Based on a state transformation, the unmeasurable states of the systems can be estimated where non-linear terms in the systems have been eliminated. It has been shown that the estimation error is linearly dynamical regarding to the presented vector-valued filter gain which can be optimised by minimising the entropy-based performance criterion. In addition, the convergence of the presented algorithm is analysed in mean-square sense and a numerical example is given to verify the effectiveness of the presented filtering algorithm. Meanwhile, the extended Kalman filter, unscented particle filter and minimum entropy filter are given for the comparisons of the filtering performance. Following the presented framework, some extensions of the presented filtering algorithm are discussed to indicate the flexibility of the filter design. The contribution of this paper can be summarised as establishing a novel minimum entropy filtering framework which consists of model transformation, entropy optimisation and convergence analysis.

42 ENGINEERING↗
DOE OSTI2010

DWPF DECON FRIT SUPERNATE ANALYSIS

The Savannah River National Laboratory (SRNL) has been requested to perform analyses on samples of the Defense Waste Processing Facility (DWPF) decon frit slurry (i.e., supernate samples and sump solid samples). Four 1-L liquid slurry samples were provided to SRNL by Savannah River Remediation (SRR) from the 'front-end' decon activities. Additionally, two 1-L sump solids samples were provided to SRNL for compositional and physical analysis. This report contains the results of the supernate analyses, while the solids (sump and slurry) results will be reported in a supplemental report. The analytical data from the decon frit supernate indicate that all of the radionuclide, organic, and inorganic concentrations met the limits in Revision 4 of the Effluent Treatment Plant (ETP) Waste Acceptance Criteria (WAC) with the exception of boron. The ETP WAC limit for boron is 15.0 mg/L while the average measured concentration (based on quadruplicate analysis) was 15.5 mg/L. The measured concentrations of Li, Na, and Si were also relatively high in the supernate analysis. These results are consistent with the relatively high measured value of B given the compositional make-up of Frit 418. Given these results, it was speculated that either (a) Frit 418 was dissolving into the supernate or aqueous fraction and/or (b) fine frit particulates were carried forward to the analytical instrument based on the sampling procedure used (i.e., the supernate samples were not filtered - only settled with the liquid fraction being transferred with a pipette). To address this issue, a filtered supernate sample (using a 0.45 um filter) was prepared and submitted for analysis. The results of the filtered sample were consistent with 'unfiltered or settled' sample - relatively high values of B, Li, Na, and Si were found. This suggests that Frit 418 is dissolving in the liquid phase which could be enhanced by the high surface area of the frit fines or particulates in suspension. Based on the results of this study, it is recommended that DWPF re-evaluate the technical basis for the B WAC limit (the only component that exceeds the ETP WAC limit from the supernate analyses) or assess if a waiver or exception can be obtained for exceeding this limit. Given the possible dissolution of B, Li, Na, and Si into the supernate (due to dissolution of frit), DWPF may need to assess if the release of these frit components into the supernate are a concern for the disposal options being considered. It should be noted that the results of this study may not be representative of future decon frit solutions or sump/slurry solids samples. Therefore, future DWPF decisions regarding the possible disposal pathways for either the aqueous or solid portions of the Decon Frit system need to factor in the potential differences. More specifically, introduction of a different frit or changes to other DWPF flowsheet unit operations (e.g., different sludge batch or coupling with other process streams) may impact not only the results but also the conclusions regarding acceptability with respect to the ETF WAC limits.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2012

A Novel Carotid Device for Embolic Diversion: Lessons Learned from a 'First in Man' Trial in Patients with Atrial Fibrillation

Purpose: This study was designed to investigate a novel device that was developed to direct emboli away from the internal carotid artery to the external carotid artery to prevent embolic stroke while avoiding the necessity for anticoagulation with warfarin. Methods: A stent-like, self-expandable filter was developed consisting of fine cobalt alloy mesh. After promising preclinical animal testing, the device was implanted into the common carotid artery extending into the external carotid artery in three patients with atrial fibrillation, each of whom were at high-risk for cerebral emboli and were suboptimal candidates for anticoagulation. Follow-up was performed clinically and with carotid duplex ultrasound. Results: The implantation was successful in all three patients. Unrestricted flow through the filter was demonstrated in each patient at 6 months by duplex sonography. However, at 7 months ultrasound demonstrated occlusion of the internal carotid artery filter in one patient, which was subsequently confirmed by angiography. Subtotal filter occlusion occurred in a second patient at 14 months. Neither patient had clinical symptoms. Both internal carotid arteries were successfully recanalized using standard angioplasty techniques without complication. The third patient had completely normal carotid blood flow at 29 months follow-up. Conclusions: Implantation of a novel device intended to divert emboli away from the internal carotid artery was feasible, but resulted in occlusive neointimal proliferation in two of three patients. These results demonstrate a discrepancy from the findings of preclinical animal studies and human investigation.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2016

TH-AB-207A-07: Radiation Dose Simulation for a Newly Proposed Dynamic Bowtie Filters for CT Using Fast Monte Carlo Methods

Purpose: Dynamic bowtie filter is an innovative design capable of modulating the X-ray and balancing the flux in the detectors, and it introduces a new way of patient-specific CT scan optimizations. This study demonstrates the feasibility of performing fast Monte Carlo dose calculation for a type of dynamic bowtie filter for cone-beam CT (Liu et al. 2014 9(7) PloS one) using MIC coprocessors. Methods: The dynamic bowtie filter in question consists of a highly attenuating bowtie component (HB) and a weakly attenuating bowtie (WB). The HB is filled with CeCl3 solution and its surface is defined by a transcendental equation. The WB is an elliptical cylinder filled with air and immersed in the HB. As the scanner rotates, the orientation of WB remains the same with the static patient. In our Monte Carlo simulation, the HB was approximated by 576 boxes. The phantom was a voxelized elliptical cylinder composed of PMMA and surrounded by air (44cm×44cm×40cm, 1000×1000×1 voxels). The dose to the PMMA phantom was tallied with 0.15% statistical uncertainty under 100 kVp source. Two Monte Carlo codes ARCHER and MCNP-6.1 were compared. Both used double-precision. Compiler flags that may trade accuracy for speed were avoided. Results: The wall time of the simulation was 25.4 seconds by ARCHER on a 5110P MIC, 40 seconds on a X5650 CPU, and 523 seconds by the multithreaded MCNP on the same CPU. The high performance of ARCHER is attributed to the parameterized geometry and vectorization of the program hotspots. Conclusion: The dynamic bowtie filter modeled in this study is able to effectively reduce the dynamic range of the detected signals for the photon-counting detectors. With appropriate software optimization methods, the accelerator-based (MIC and GPU) Monte Carlo dose engines have shown good performance and can contribute to patient-specific CT scan optimizations.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2014

Air Sample Conditioner Helps the Waste Treatment Plant Meet Emissions Standards

The air in three of the Hanford Site Waste Treatment and Immobilization Plant (WTP) melter off-gas discharge stacks will be hot and humid after passing through the train of emission abatement equipment. The off-gas temperature and humidity levels will be incompatible with the airborne emissions monitoring equipment required for this type of stack. To facilitate sampling from these facilities, an air sample conditioner system will be installed to introduce cool, dry air into the sample stream to reduce the temperature and dew point. This will avoid thermal damage to the instrumentation and problematic condensation. The complete sample transport system must also deliver at least 50% of the particles in the sample airstream to the sample collection and on-line analysis equipment. The primary components of the sample conditioning system were tested in a laboratory setting. The sample conditioner itself is based on a commercially-available porous tube filter design. It consists of a porous sintered metal tube inside a coaxial metal jacket. The hot gas sample stream passes axially through the porous tube, and the dry, cool air is injected into the jacket and through the porous wall of the inner tube, creating an effective sample diluter. The dilution and sample air mix along the entire length of the porous tube, thereby simultaneously reducing the dew point and temperature of the mixed sample stream. Furthermore, because the dilution air enters through the porous tube wall, the sample stream does not come in contact with the porous wall and particle deposition is reduced in this part of the sampling system. Tests were performed with an environmental chamber to supply air with the temperature and humidity needed to simulate the off-gas conditions. Air from the chamber was passed through the conditioning system to test its ability to reduce the temperature and dew point of the sample stream. To measure particle deposition, oil droplets in the range of 9 to 11 micrometer aerodynamic diameter were injected into the environmental chamber and drawn through the conditioning system, which included a filter to capture droplets that passed through the conditioner. The droplets were tagged with a fluorescent dye which allowed quantification of droplet deposition on each component of the system. The tests demonstrated the required reductions in temperature and moisture, with no condensation forming when heat tracing was added on the upstream end of the sample conditioner. Additionally, tests indicated that the system, operating at several flow rates and in both vertical and horizontal orientations, delivers nearly all of the sampled particles for analysis. Typical aerosol penetration values were between 98 and 99%. PNNL, Bechtel National Inc., and the instrument vendor are working to implement the sample conditioner into the air monitoring systems used for the melter off-gas exhaust streams. Similar technology may be useful for processes in other facilities with air exhaust streams with elevated temperature and/or humidity.

sample conditioning↗
DOE OSTI2016

Model predictive direct power control for active power decoupled single-phase quasi- Z -source inverter

In this study, the active power filter (APF) that consists of a half-bridge leg and an ac capacitor is integrated in the single-phase quasi-Z-source inverter (qZSI) in this paper to avoid the second harmonic power flowing into the dc side. The capacitor of APF buffers the second harmonic power of the load, and the ac capacitor allows highly pulsating ac voltage, so that the capacitances of both dc and ac sides can be small. A model predictive direct power control (DPC) is further proposed to achieve the purpose of this newtopology through predicting the capacitor voltage of APF at each sampling period and ensuring the APF power to track the second harmonic power of single-phase qZSI. Simulation and experimental results verify the model predictive DPC for the APF-integrated single-phase qZSI.

30 DIRECT ENERGY CONVERSION↗
DOE OSTI2016

Effect of Organic Coatings, Humidity and Aerosol Acidity on Multiphase Chemistry of Isoprene Epoxydiols

Multiphase chemistry of isomeric isoprene epoxydiols (IEPOX) has been shown to be the dominant source of isoprene-derived secondary organic aerosol (SOA). Recent studies have reported particles composed of ammonium bisulfate (ABS) mixed with model organics exhibit slower rates of IEPOX uptake. In the present study, we investigate the effect of atmospherically-relevant organic coatings of α-pinene (AP) SOA on the reactive uptake of trans-β-IEPOX onto ABS particles under different conditions and coating thicknesses. Single particle mass spectrometry was used to characterize in real-time particle size, shape, density, and quantitative composition before and after reaction with IEPOX. We find that IEPOX uptake by pure sulfate particles is a volume-controlled process, which results in particles with uniform concentration of IEPOX-derived SOA across a wide range of sizes. Aerosol acidity was shown to enhance IEPOX-derived SOA formation, consistent with recent studies. The presence of water has a weaker impact on IEPOX-derived SOA yield, but significantly enhanced formation of 2-methyltetrols, consistent with offline filter analysis. In contrast, IEPOX uptake by ABS particles coated by AP-derived SOA is strongly dependent on particle size and composition. IEPOX uptake occurred only when weight fraction of AP-derived SOA dropped below 50 %, effectively limiting IEPOX uptake to larger particles.

Environmental Molecular Sciences Laboratory↗
DOE OSTI2021

Strategies toward the Synthesis of Advanced Functional Sorbent Performance for Uranium Uptake from Seawater

A polymer fiber-based adsorbent (AF1) composed of acrylonitrile and itaconic acid functional groups was synthesized by a radiation-induced graft polymerization technique onto hollow-gear shaped polyethylene (PE) fibers. Investigation of the optimum reaction parameters for the conversion of grafted cyano moieties into amidoxime groups was conducted by the reaction with hydroxylamine at different temperatures and periods in a variety of aqueous and organic solvents. The 13 C CP/MAS spectra of amidoximated AF1 adsorbent fibers from water–methanol and dimethyl sulfoxide (DMSO) revealed that both the cyclic imide dioxime and open-chain amidoxime were formed through the reaction. The conversion from amidoxime to imide dioxime was found to occur slowly and gradually with the increasing reaction time. The quicker diffusion of DMSO as compared to that of water–methanol, in the grafted trunk PE fiber, resulted in faster kinetics of the amidoximation reaction. The uranium adsorption capacity of the amidoximated AF1 samples was determined after (i) 24 h contact with sodium-based brine spiked with uranium and (ii) 56 days of exposure to filtered seawater (Sequim Bay, WA, USA) in flow-through columns. Finally, the uranium extraction performance of the adsorbents after exposure to filtered seawater was consistent with the laboratory screening results, and the amidoximated AF1 samples (in DMSO at 70 °C for 3 h) exhibited the highest 56 day uranium adsorption capacity (5.04 ± 0.15 g U/kg-ads) with faster adsorption kinetics compared to the pristine AF1 adsorbents.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2014

Electrical detection of spin hyperpolarization in InP

The electrical detection of surface spin polarization in Indium Phosphide (InP) is demonstrated. Using a planar four-terminal architecture on top of semi-insulating Fe:InP (001) wafers, optical orientation is separated from electrical detection. Spin filter tunnel contacts consisting of InP/oxide/Co reveal significant asymmetries in the differential resistance upon helicity change of the optical pumping. The iron-rich tunnel oxide provides the main spin selection mechanism. A reproducible helicity-dependent asymmetry as high as 18% could be observed at T = 55 K and an external induction field μ{sub 0}H = 1 T. At room temperature and zero external field, a helicity-dependent asymmetry of 6% suggests the stand-alone applicability of the device either as an electronic spin sensor or as an optical helicity sensor.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2015

Gain and frequency tuning within the mouse cochlear apex

Normal mammalian hearing requires cochlear outer hair cell active processes that amplify the traveling wave with high gain and sharp tuning, termed cochlear amplification. We have used optical coherence tomography to study cochlear amplification within the apical turn of the mouse cochlea. We measured not only classical basilar membrane vibratory tuning curves but also vibratory responses from the rest of the tissues that compose the organ of Corti. Basilar membrane tuning was sharp in live mice and broad in dead mice, whereas other regions of the organ of Corti demonstrated phase shifts consistent with additional filtering beyond that provided by basilar membrane mechanics. We use these experimental data to support a conceptual framework of how cochlear amplification is tuned within the mouse cochlear apex. We will also study transgenic mice with targeted mutations that affect different biomechanical aspects of the organ of Corti in an effort to localize the underlying processes that produce this additional filtering.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2026

Phase Picking Beyond Local Distances: Where Waveform Filtering Still Matters for Deep Learning Models

Waveform filtering is a standard step in traditional seismic phase picking but often receives little attention in deep learning workflows, where models are typically trained on raw or minimally processed waveforms. Although this strategy performs well for local events, we show that performance can degrade substantially at regional distances. To address this limitation, we introduce two ways to incorporate multiband-filtered waveforms into deep learning phase pickers. The stacking approach concatenates filtered inputs along the channel dimension, while the branching approach processes each frequency band through a dedicated network branch before feature fusion. Both approaches can substantially improve performance across epicentral distances of 0° to 20°, but their effectiveness depends strongly on the selected frequency bands. Tests with multiple filter banks show that filter-bank design should be treated as part of model optimization rather than as a fixed preprocessing choice. Grad-CAM analysis of the branching model indicates that band importance varies among waveform samples and across training realizations, with only a weak overall preference for the 0.25 to 0.5 Hz band. These results show that no single filter band is consistently optimal and demonstrate that explicit feature engineering remains valuable for robust deep learning-based seismic phase picking.

58 GEOSCIENCES↗
DOE OSTI2016

Interim Status of HEPA Filter 10-Year Lifetime Evaluation

High efficiency particulate air (HEPA) filters are used to remove a variety of particles from industrial gas streams. HEPA filters remove at least 99.97% of airborne particles 0.3 microns in diameter, while particles smaller or larger than 0.3 microns are removed with an even higher efficiency. The purpose of this study is to draw a relationship between the HEPA filter lifespan and various competing factors that include the differential pressure drop in nuclear grade HEPA filters, the fume hood face velocity, the radiological dose, and the filter efficiency characteristics. The Department of Energy (DOE) uses a conservative interpretation of data to set the age limit of HEPA filters at 10 years (DOE-HDBK-1169-2003). The lifetime was determined by an analysis of multiple HEPA filter research studies, performed by Bergman at the Lawrence Livermore National Laboratory (LLNL). Analysis of data from Robinson et al. suggests that unfolded media tensile strength fail at 13 years, while folded media do not have the required 2.5 pound/inch tensile strength even when new and is extremely low at 7 years. Therefore, the data displayed failed tensile strength and low burst strengths at an average of 10 years. Although filter life was difficult to estimate using the data, based on the relationship between HEPA filter and their age, the recommended lifetime was set at 10 years under dry conditions (Bergman 1999). Effluent Management (EM) is coordinating a 10-year effort to collect and analyze the lifetime of HEPA filters against the other competing requirements. The study began in 2010 and consists of 49 HEPA filters that are analyzed annually to verify compliance with permit conditions. The data are collected from the Physical Science Facility (PSF) 3410, 3420, and 3430 buildings located at the Pacific Northwest National Laboratory (PNNL). The study suggests the frequency of filter replacement should also be based on the differential pressure, filter efficiency, and/or face velocity of the fume hood, rather than a prescribed filter “lifetime”, which is currently every 10 years from the date of manufacture for filters operating under dry conditions. Over the past six years, only two filters in the study were replaced: one due to high DP and low fume hood face velocity, and the other due to low filter efficiency results.

54 ENVIRONMENTAL SCIENCES↗