Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Performance Rating Method”

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 1,585 records · Page 88
All sourcesNASA NTRSDOE OSTI
DOE OSTI2016

Cost-Effectiveness Analysis of Intensity Modulated Radiation Therapy Versus 3-Dimensional Conformal Radiation Therapy for Preoperative Treatment of Extremity Soft Tissue Sarcomas

Purpose: Create a cost-effectiveness model comparing preoperative intensity modulated radiation therapy (IMRT) versus 3-dimensional conformal radiation therapy (3DCRT) for extremity soft tissue sarcomas. Methods and Materials: Input parameters included 5-year local recurrence rates, rates of acute wound adverse events, and chronic toxicities (edema, fracture, joint stiffness, and fibrosis). Health-state utilities were used to calculate quality-adjusted life years (QALYs). Overall treatment costs per QALY or incremental cost-effectiveness ratio (ICER) were calculated. Roll-back analysis was performed using average costs and utilities to determine the baseline preferred radiation technique. One-way, 2-way, and probabilistic sensitivity analyses (PSA) were performed for input parameters with the largest impact on the ICER. Results: Overall treatment costs were $17,515.58 for 3DCRT compared with $22,920.51 for IMRT. The effectiveness was higher for IMRT (3.68 QALYs) than for 3DCRT (3.35 QALYs). The baseline ICER for IMRT was $16,842.75/QALY, making it the preferable treatment. The ICER was most sensitive to the probability of local recurrence, upfront radiation costs, local recurrence costs, certain utilities (no toxicity/no recurrence, grade 1 toxicity/no local recurrence, grade 4 toxicity/no local recurrence), and life expectancy. Dominance patterns emerged when the cost of 3DCRT exceeded $15,532.05 (IMRT dominates) or the life expectancy was under 1.68 years (3DCRT dominates). Furthermore, preference patterns changed based on the rate of local recurrence (threshold: 13%). The PSA results demonstrated that IMRT was the preferred cost-effective technique for 64% of trials compared with 36% for 3DCRT. Conclusions: Based on our model, IMRT is the preferred technique by lowering rates of local recurrence, severe toxicities, and improving QALYs. From a third-party payer perspective, IMRT should be a supported approach for extremity soft tissue sarcomas.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2018

Synergistic effect of hierarchical SnO2 nanorods/Fe2O3 hexahedrons with enhanced performance as lithium ion battery anodes

Highlights: • SnO{sub 2}-Fe{sub 2}O{sub 3} nanocomposites were fabricated by a two-step hydrothermal method. • SnO{sub 2}-Fe{sub 2}O{sub 3} nanocomposites show high capacity and excellent cycling stability. • The enhanced performances benefit from the synergistic effect of SnO{sub 2} and Fe{sub 2}O{sub 3}. • The reduced Fe nanoparticles facilitate the fully conversion reaction of SnO{sub 2}. - Abstract: Hierarchical SnO{sub 2}-Fe{sub 2}O{sub 3} nanocomposites composed of SnO{sub 2} nanorods and Fe{sub 2}O{sub 3} hexahedrons are successfully prepared by a two-step hydrothermal method. The lithium ion batteries (LIBs) based on as-synthesized products are fabricated to investigate their electrochemical performance. As an anode material for LIBs, SnO{sub 2}-Fe{sub 2}O{sub 3} nanocomposites deliver a high capacity of 1022 mAh g{sup −1} at 100 mA g{sup −1} after 100 cycles. The cycling stability and rate performance of the nanocomposites are significantly improved. The enhanced performances benefit from the synergistic effect of SnO{sub 2} nanorods and Fe{sub 2}O{sub 3} hexahedrons. The iron (Fe) nanoparticles reduced from Fe{sub 2}O{sub 3} not only facilitate the conversion reaction of SnO{sub 2}, resulting in the high capacity, but also increase the conductivity of the material. In addition, SnO{sub 2} serves as a matrix to protect Fe{sub 2}O{sub 3} from agglomeration, and the hierarchical structure effectively buffers volume changes, thus enhancing the cycling life of LIBs.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2025

Formation trajectories of solution-processed perovskite thin films from mixed solvents

The engineering of mixed-solvent formulations and their evaporation conditions are key to reproducible perovskite coatings for high-performance photovoltaics. Here, we report a lumped-parameter evaporation model to predict the evolution of a perovskite ink liquid film over time (solvent ratio, solute concentration, and film thickness). The drying-rate model is validated via in situ film-thickness measurements, and the predicted transient liquid film state is mapped as a process path. These methods allow for the prediction of process sensitivity to local environmental factors and the understanding and visualization of a broader processing parameter space enabled through the coupling of process and ink engineering. Process maps are applied to create a new framework for scalable perovskite coating development with a goal of improving the reproducibility and transferability of perovskite fabrication. This approach is demonstrated with blade-coated FA 0.83 Cs 0.17 PbI 3 photovoltaic devices, improving the photovoltaic conversion efficiency from 17.5% ± 1.7% to 20.3% ± 0.6%.

14 SOLAR ENERGY↗
DOE OSTI2014

Fixed-target protein serial microcrystallography with an x-ray free electron laser

We present results from experiments at the Linac Coherent Light Source (LCLS) demonstrating that serial femtosecond crystallography (SFX) can be performed to high resolution (,2.5 Å) using protein microcrystals deposited on an ultra-thin silicon nitride membrane and embedded in a preservation medium at room temperature. Data can be acquired at a high acquisition rate using x-ray free electron laser sources to overcome radiation damage, while sample consumption is dramatically reduced compared to flowing jet methods. We achieved a peak data acquisition rate of 10 Hz with a hit rate of ,38%, indicating that a complete data set could be acquired in about one 12-hour LCLS shift using the setup described here, or in even less time using hardware optimized for fixed target SFX. This demonstration opens the door to ultra low sample consumption SFX using the technique of diffraction-before-destruction on proteins that exist in only small quantities and/or do not produce the copious quantities of microcrystals required for flowing jet methods.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2017

CMU DeepLens: deep learning for automatic image-based galaxy–galaxy strong lens finding

Galaxy-scale strong gravitational lensing can not only provide a valuable probe of the dark matter distribution of massive galaxies, but also provide valuable cosmological constraints, either by studying the population of strong lenses or by measuring time delays in lensed quasars. Due to the rarity of galaxy-scale strongly lensed systems, fast and reliable automated lens finding methods will be essential in the era of large surveys such as Large Synoptic Survey Telescope, Euclid and Wide-Field Infrared Survey Telescope. To tackle this challenge, we introduce CMU DeepLens, a new fully automated galaxy-galaxy lens finding method based on deep learning. This supervised machine learning approach does not require any tuning after the training step which only requires realistic image simulations of strongly lensed systems. We train and validate our model on a set of 20 000 LSST-like mock observations including a range of lensed systems of various sizes and signal-to-noise ratios (S/N). We find on our simulated data set that for a rejection rate of non-lenses of 99 per cent, a completeness of 90 per cent can be achieved for lenses with Einstein radii larger than 1.4 arcsec and S/N larger than 20 on individual g-band LSST exposures. Finally, we emphasize the importance of realistically complex simulations for training such machine learning methods by demonstrating that the performance of models of significantly different complexities cannot be distinguished on simpler simulations. We make our code publicly available at https://github.com/McWilliamsCenter/CMUDeepLens.

gravitational lensing: strong↗
DOE OSTI2012

Nanoflow electrospinning serial femtosecond crystallography

A low flow rate liquid microjet method for delivery of hydrated protein crystals to X-ray lasers is presented. Linac Coherent Light Source data demonstrates serial femtosecond protein crystallography with micrograms, a reduction of sample consumption by orders of magnitude. An electrospun liquid microjet has been developed that delivers protein microcrystal suspensions at flow rates of 0.14–3.1 µl min{sup −1} to perform serial femtosecond crystallography (SFX) studies with X-ray lasers. Thermolysin microcrystals flowed at 0.17 µl min{sup −1} and diffracted to beyond 4 Å resolution, producing 14 000 indexable diffraction patterns, or four per second, from 140 µg of protein. Nanoflow electrospinning extends SFX to biological samples that necessitate minimal sample consumption.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2021

Assessing Machine Learning as a Tool to Explain Variance in Deployed Photovoltaic (PV) System Degradation

Degradation remains a large uncertainty in forecasting production for PV plants, creating significant risk for developers and financiers. This study aims to quantify the distribution and drivers of degradation across 10,000 PV systems deployed for distributed or utility generation by training a machine learning model to predict year-over-year degradation rates from metadata characteristics. A combination of K-Means clustering and random forest regressor were found to associate multiple metadata features as potential drivers of degradation, including module characteristics, system design, and climate features. From this, it is inferred that if machine learning is able to find complex patterns between metadata features and system performance loss, such methods can be employed to help developers and financiers make data-informed decisions when estimating long-term energy production forecasts in financial models.

Dunn, Jimmy C.↗
DOE OSTI2013

Characterization of a synthetic single crystal diamond Schottky diode for radiotherapy electron beam dosimetry

Purpose: To investigate the dosimetric properties of synthetic single crystal diamond based Schottky diodes under irradiation with therapeutic electron beams from linear accelerators. Methods: A single crystal diamond detector was fabricated and tested under 6, 8, 10, 12, and 15 MeV electron beams. The detector performances were evaluated using three types of commercial detectors as reference dosimeters: an Advanced Markus plane parallel ionization chamber, a Semiflex cylindrical ionization chamber, and a p-type silicon detector. Preirradiation, linearity with dose, dose rate dependence, output factors, lateral field profiles, and percentage depth dose profiles were investigated and discussed. Results: During preirradiation the diamond detector signal shows a weak decrease within 0.7% with respect to the plateau value and a final signal stability of 0.1% (1{sigma}) is observed after about 5 Gy. A good linear behavior of the detector response as a function of the delivered dose is observed with deviations below {+-}0.3% in the dose range from 0.02 to 10 Gy. In addition, the detector response is dose rate independent, with deviations below 0.3% in the investigated dose rate range from 0.17 to 5.45 Gy/min. Percentage depth dose curves obtained from the diamond detector are in good agreement with the ones from the reference dosimeters. Lateral beam profile measurements show an overall good agreement among detectors, taking into account their respective geometrical features. The spatial resolution of solid state detectors is confirmed to be better than that of ionization chambers, being the one from the diamond detector comparable to that of the silicon diode. A good agreement within experimental uncertainties was also found in terms of output factor measurements between the diamond detector and reference dosimeters. Conclusions: The observed dosimetric properties indicate that the tested diamond detector is a suitable candidate for clinical electron beam dosimetry.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2014

TU-C-BRE-08: IMRT QA: Selecting Meaningful Gamma Criteria Based On Error Detection Sensitivity

Purpose: To develop a strategy for defining meaningful tolerance limits and studying the sensitivity of IMRT QA gamma criteria by inducing known errors in QA plans. Methods: IMRT QA measurements (ArcCHECK, Sun Nuclear) were compared to QA plan calculations with induced errors. Many (>24) gamma comparisons between data and calculations were performed for each of several kinds of cases and classes of induced error types with varying magnitudes (e.g. MU errors ranging from -10% to +10%), resulting in over 3,000 comparisons. Gamma passing rates for each error class and case were graphed against error magnitude to create error curves in order to represent the range of missed errors in routine IMRT QA using various gamma criteria. Results: This study demonstrates that random, case-specific, and systematic errors can be detected by the error curve analysis. Depending on location of the peak of the error curve (e.g., not centered about zero), 3%/3mm threshold=10% criteria may miss MU errors of up to 10% and random MLC errors of up to 5 mm. Additionally, using larger dose thresholds for specific devices may increase error sensitivity (for the same X%/Ymm criteria) by up to a factor of two. This analysis will allow clinics to select more meaningful gamma criteria based on QA device, treatment techniques, and acceptable error tolerances. Conclusion: We propose a strategy for selecting gamma parameters based on the sensitivity of gamma criteria and individual QA devices to induced calculation errors in QA plans. Our data suggest large errors may be missed using conventional gamma criteria and that using stricter criteria with an increased dose threshold may reduce the range of missed errors. This approach allows quantification of gamma criteria sensitivity and is straightforward to apply to other combinations of devices and treatment techniques.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2015

A method for modeling laterally asymmetric proton beamlets resulting from collimation

Purpose: To introduce a method to model the 3D dose distribution of laterally asymmetric proton beamlets resulting from collimation. The model enables rapid beamlet calculation for spot scanning (SS) delivery using a novel penumbra-reducing dynamic collimation system (DCS) with two pairs of trimmers oriented perpendicular to each other. Methods: Trimmed beamlet dose distributions in water were simulated with MCNPX and the collimating effects noted in the simulations were validated by experimental measurement. The simulated beamlets were modeled analytically using integral depth dose curves along with an asymmetric Gaussian function to represent fluence in the beam’s eye view (BEV). The BEV parameters consisted of Gaussian standard deviations (sigmas) along each primary axis (σ{sub x1},σ{sub x2},σ{sub y1},σ{sub y2}) together with the spatial location of the maximum dose (μ{sub x},μ{sub y}). Percent depth dose variation with trimmer position was accounted for with a depth-dependent correction function. Beamlet growth with depth was accounted for by combining the in-air divergence with Hong’s fit of the Highland approximation along each axis in the BEV. Results: The beamlet model showed excellent agreement with the Monte Carlo simulation data used as a benchmark. The overall passing rate for a 3D gamma test with 3%/3 mm passing criteria was 96.1% between the analytical model and Monte Carlo data in an example treatment plan. Conclusions: The analytical model is capable of accurately representing individual asymmetric beamlets resulting from use of the DCS. This method enables integration of the DCS into a treatment planning system to perform dose computation in patient datasets. The method could be generalized for use with any SS collimation system in which blades, leaves, or trimmers are used to laterally sharpen beamlets.

07 ISOTOPE AND RADIATION SOURCES↗
DOE OSTI2016

SU-F-T-301: Planar Dose Pass Rate Inflation Due to the MapCHECK Measurement Uncertainty Function

Purpose: To quantify the effect of the Measurement Uncertainty function on planar dosimetry pass rates, as analyzed with Sun Nuclear Corporation analytic software (“MapCHECK” or “SNC Patient”). This optional function is toggled on by default upon software installation, and automatically increases the user-defined dose percent difference (%Diff) tolerance for each planar dose comparison. Methods: Dose planes from 109 IMRT fields and 40 VMAT arcs were measured with the MapCHECK 2 diode array, and compared to calculated planes from a commercial treatment planning system. Pass rates were calculated within the SNC analytic software using varying calculation parameters, including Measurement Uncertainty on and off. By varying the %Diff criterion for each dose comparison performed with Measurement Uncertainty turned off, an effective %Diff criterion was defined for each field/arc corresponding to the pass rate achieved with MapCHECK Uncertainty turned on. Results: For 3%/3mm analysis, the Measurement Uncertainty function increases the user-defined %Diff by 0.8–1.1% average, depending on plan type and calculation technique, for an average pass rate increase of 1.0–3.5% (maximum +8.7%). For 2%, 2 mm analysis, the Measurement Uncertainty function increases the user-defined %Diff by 0.7–1.2% average, for an average pass rate increase of 3.5–8.1% (maximum +14.2%). The largest increases in pass rate are generally seen with poorly-matched planar dose comparisons; the MapCHECK Uncertainty effect is markedly smaller as pass rates approach 100%. Conclusion: The Measurement Uncertainty function may substantially inflate planar dose comparison pass rates for typical IMRT and VMAT planes. The types of uncertainties incorporated into the function (and their associated quantitative estimates) as described in the software user’s manual may not accurately estimate realistic measurement uncertainty for the user’s measurement conditions. Pass rates listed in published reports or otherwise compared to the results of other users or vendors should clearly indicate whether the Measurement Uncertainty function is used.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Using spatio-temporal graph neural networks to estimate fleet-wide photovoltaic performance degradation patterns

Accurate estimation of photovoltaic (PV) system performance is crucial for determining its feasibility as a power generation technology and financial asset. PV-based energy solutions offer a viable alternative to traditional energy resources due to their superior Levelized Cost of Energy (LCOE). A significant challenge in assessing the LCOE of PV systems lies in understanding the Performance Loss Rate (PLR) for large fleets of PV systems. Estimating the PLR of PV systems becomes increasingly important in the rapidly growing PV industry. Precise PLR estimation benefits PV users by providing real-time monitoring of PV module performance, while explainable PLR estimation assists PV manufacturers in studying and enhancing the performance of their products. However, traditional PLR estimation methods based on statistical models have notable drawbacks. Firstly, they require user knowledge and decision-making. Secondly, they fail to leverage spatial coherence for fleet-level analysis. Additionally, these methods inherently assume the linearity of degradation, which is not representative of real world degradation. To overcome these challenges, we propose a novel graph deep learning-based decomposition method called the Spatio-Temporal Graph Neural Network for fleet-level PLR estimation (PV-stGNN-PLR). PV-stGNN-PLR decomposes the power timeseries data into aging and fluctuation components, utilizing the aging component to estimate PLR. PV-stGNN-PLR exploits spatial and temporal coherence to derive PLR estimation for all systems in a fleet and imposes flatness and smoothness regularization in loss function to ensure the successful disentanglement between aging and fluctuation. We have evaluated PV-stGNN-PLR on three simulated PV datasets consisting of 100 inverters from 5 sites. Experimental results show that PV-stGNN-PLR obtains a reduction of 33.9% and 35.1% on average in Mean Absolute Percent Error (MAPE) and Euclidean Distance (ED) in PLR degradation pattern estimation compared to the state-of-the-art PLR estimation methods.

14 SOLAR ENERGY↗
DOE OSTI2020

A Pilot-Scale Evaluation of Natural Gas-Based Foam at Elevated Pressure and Temperature Conditions

Abstract Nitrogen (N2) and Carbon Dioxide (CO2) foams have been used as hydraulic fracturing fluids for several decades to reduce water usage and minimize damage in water-sensitive reservoirs. These foam treatments require gases to be liquefied and transported to site. An alternative approach would be to use natural gas (NG) that is readily available from nearby wells, pipelines, and processing facilities as the internal, gaseous phase to create a NG-based foam. Hydraulic fracturing with NG foam is a relatively inexpensive option, makes use of an abundant and often wasted resource, and may even provide production benefits in certain reservoirs. As part of an ongoing development project sponsored by the Department of Energy (DOE), the surface process to create NG foam is being developed and the properties of NG foam are being explored. This paper presents recent results from a rigorous pilot-scale demonstration of NG foam over a range of operating scenarios relevant to surface and bottomhole conditions with a variety of base-fluid mixtures. The Pilot-scale Foam Test Facility (PFTF) used in these investigations is first described. The PFTF is capable of generating foamed fluids at pressures up to 7,500 psig and at temperatures in excess of 300°F. Then, results from several investigations aimed at proving NG foam at conditions relevant to the field are presented. NG foam was characterized using rheology measurements and flow visualization techniques. Experiments were performed to investigate the texture and stability of NG foam generated by two different mixing methods: one using a custom designed tee to match mixing velocities in the field where the gas phase is jetted into the aqueous stream, and another to ensure comprehensive mixing for laboratory analysis. Parametric studies were conducted to explore the effects of flow rate, foam quality, and temperature on the stability of NG foam. Moreover, different fluid preparations were used to investigate the effect of base fluid and additive concentrations on the stability of NG foams. Additional laboratory studies that investigated foam stability with produced water and multicomponent NG mixtures are also reported. The NG foams explored in these investigations exhibited typical, shear-thinning behavior observed in rheological studies of N2- and CO2-based foams. The measured viscosity and observed stability indicate that NG foams are well suited for fracturing applications. Like other foams, NG foam exhibits sensitivity to operating temperature characterized by a decrease in apparent viscosity as temperature increases. Rapid foam breakdown was observed at significantly elevated temperatures exceeding 290°F. In addition to fluid characterization, these investigations also yielded several key lessons that should be applied to future field demonstrations of NG foam.

Beck, Griffin↗
DOE OSTI2021

Acetonitrile Destruction and Fate of Organics in the Reverse Osmosis System at the ETF

The Hanford Site Effluent Treatment Facility (ETF) currently treats aqueous waste streams that include condensates from the 242-A evaporator, leachate from the Environmental Restoration Disposal Facility (ERDF), as well as laboratory wastes and, in the future, will treat liquid effluents from the Hanford Tank Waste Treatment and Immobilization Plant (WTP) and Integrated Disposal Facility (IDF) leachate. Liquid effluents from the WTP will have significant concentrations of acetonitrile. Acetonitrile is formed by reaction of nitrates and sugar in the WTP low activity waste (LAW) melters and is prevalent in the submerged bed scrubber (SBS) and wet electrostatic precipitator (WESP) liquid effluents from WTP off-gas treatment. When these liquids are concentrated in the WTP Effluent Management Facility (EMF) evaporator in the direct feed low activity waste (DFLAW) flow-sheet, testing has shown that the majority of the acetonitrile partitions to the evaporator condensate. Since the evaporator condensate is directed to the ETF, this creates a potential issue with the ETF waste acceptance criteria. Consequently, there is a need to validate flow-sheet assumptions on the fate of acetonitrile and other organics within the ETF. The present plan includes the addition of a steam stripper to the ETF to remove acetonitrile. There is, therefore, also a need to determine a suitable method to destroy acetonitrile in the overhead condensate stream from the new steam stripper. Washington River Protection Solutions, LLC (WRPS) previously contracted with Atkins and the Vitreous State Laboratory (VSL) of The Catholic University of America (CUA) to perform development and testing work to evaluate potential methods for destruction of acetonitrile in WTP secondary liquid effluents. Based on the results of that work, WRPS requested that follow-on testing be conducted to further evaluate acetonitrile destruction in the steam stripper condensate using ultraviolet oxidation (UV/OX) with persulfate. WRPS also requested testing to assess the rejection rate of organics in the reverse osmosis (RO) system installed in the ETF. This report presents the results from testing to address those needs.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2020

CSB-RNN: A Faster-Than-Realtime RNN Acceleration Framework with Compressed Structured Blocks

Recurrent Neural Networks (RNN) is widely applied to temporal sequence analysis, where real-time performance is usually in demand. However, RNN suffers a heavy computational workload as the model comes with a large weight matrix. To alleviate the pain, model compression (pruning) schemes have been proposed for RNN that pruning the redundant (near-zero) weight-values. On the one hand, the non-structured pruning methods achieve a considerable pruning rate while bringing the computational irregularity, which is un-friendly to parallel-hardware. On the other hand, the existing structured pruning methods consider the hardware parallelism; However, they suffer a poor pruning rate due to the restrict constraints on pruning structure. This paper presents CSB-RNN, an optimized full-stack RNN framework with the novel compressed structured block (CSB) technique. The CSB-pruned RNN model comes with both fine-granularity that benefits the pruning rate and regular structure that facilitates the hardware-parallelism. Further, we propose a novel hardware architecture for inferencing the CSB-pruned model. Different from conventional parallel hardware, this architecture solves the block-workload imbalance issue and achieves an over 95% hardware utilization. With the experiments on 10 RNN models in 5 application domains, the CSB-RNN realizes 7×-20× lossless compression and up to 50× acceptable lossy-compression, which is 2×-7× to the prior art. With the addition of the novel hardware, the compressed-RNN inference reaches a super real-time latency of 10-400µs with FPGA implementation.

Shi, Runbin↗
DOE OSTI2020

NSFnets (Navier-Stokes flow nets): Physics-informed neural networks for the incompressible Navier-Stokes equations

In the last 50 years there has been a tremendous progress in solving numerically the Navier-Stokes equations using finite differences, finite elements, spectral, and even meshless methods. Yet, in many real cases, we still cannot incorporate seamlessly (multi-fidelity) data into existing algorithms, and for industrial-complexity applications the mesh generation is time consuming and still an art. Moreover, solving ill-posed problems (e.g., lacking boundary conditions) or inverse problems is often prohibitively expensive and requires different formulations and new computer codes. Here, we employ physics-informed neural networks (PINNs), encoding the governing equations directly into the deep neural network via automatic differentiation, to overcome some of the aforementioned limitations for simulating incompressible laminar and turbulent flows. We develop the Navier-Stokes flow nets (NSFnets) by considering two different mathematical formulations of the Navier-Stokes equations: the velocity-pressure (VP) formulation and the vorticity-velocity (VV) formulation. Since this is a new approach, we first select some standard benchmark problems to assess the accuracy, convergence rate, computational cost and flexibility of NSFnets; analytical solutions and direct numerical simulation (DNS) databases provide proper initial and boundary conditions for the NSFnet simulations. The spatial and temporal coordinates are the inputs of the NSFnets, while the instantaneous velocity and pressure fields are the outputs for the VP-NSFnet, and the instantaneous velocity and vorticity fields are the outputs for the VV-NSFnet. This is unsupervised learning and, hence, no labeled data are required beyond boundary and initial conditions and the fluid properties. The residuals of the VP or VV governing equations, together with the initial and boundary conditions, are embedded into the loss function of the NSFnets. No data is provided for the pressure to the VP-NSFnet, which is a hidden state and is obtained via the incompressibility constraint without extra computational cost. Unlike the traditional numerical methods, NSFnets inherit the properties of neural networks (NNs), hence the total error is composed of the approximation, the optimization, and the generalization errors. Here, we empirically attempt to quantify these errors by varying the sampling (“residual”) points, the iterative solvers, and the size of the NN architecture. For the laminar flow solutions, we show that both the VP and the VV formulations are comparable in accuracy but their best performance corresponds to different NN architectures. The initial convergence rate is fast but the error eventually saturates to a plateau due to the dominance of the optimization error. For the turbulent channel flow, we show that NSFnets can sustain turbulence at , but due to expensive training we only consider part of the channel domain and enforce velocity boundary conditions on the subdomain boundaries provided by the DNS data base. We also perform a systematic study on the weights used in the loss function for balancing the data and physics components, and investigate a new way of computing the weights dynamically to accelerate training and enhance accuracy. In the last part, we demonstrate how NSFnets should be used in practice, namely for ill-posed problems with incomplete or noisy boundary conditions as well as for inverse problems. We obtain reasonably accurate solutions for such cases as well without the need to change the NSFnets and at the same computational cost as in the forward well-posed problems. As a result, we also present a simple example of transfer learning that will aid in accelerating the training of NSFnets for different parameter settings.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

SU-F-T-550: Radiochromic Plastic Thin Sheet Dosimeter: Initial Performance

Purpose: Thin sheets, of a high sensitivity formulation of radiochromic dosimeter, Presage were prepared and evaluated for optical readout. Methods: Sheets of radiochromic polyurethane, 12 cm long, 10 cm wide and 0.2 cm thick were prepared with leuco crystal violet as the reporter molecule. Sample transmission was evaluated at a wavelength of 590 nm with in-house constructed instruments: optical cone beam laser CT scanner, fixed and scanning spot densitometers. Sample sequential irradiations to a total dose of 40 Gy were conducted with a modified, Theratron 60, cobalt radiotherapy machine at dose rates of 1 or 0.25 Gy per minute. Exposure to ambient and readout light was minimized to limit background photochromic signals. Samples were stored at 4°C. Optical activity was assessed from linearly polarized transmission images. Comparison sensitivity measurements with EBT3 film were conducted. Results: Samples were transparent, smooth and pale purple before irradiation. Radiochromic reaction was completed in less than 5 minutes. A linear dose response with a sensitivity of 0.5 cm-1Gy-1 was observed. Micrometer measurements found sheet thickness variations up to 20%. Uniform dose, 2 Gy attenuation images, correlated with local sheet thicknesses. Comparable measurements with EBT3 film were 3 times more sensitive at 1 Gy but above 15 Gy, EBT3 film had lower sensitivity than 0.2 cm thick Presage sheet dosimeter due to its non-linear response. Conclusion: Dose sensitivity provided a 10% decrease in transmission for a 1 Gy dose. Improvements in mold design are expected to allow production of sheets with less than 5% variation in thickness. Above, 10 Gy, Presage sheet dosimeter performance expected to exceed EBT3 film based on linearity, sensitivity, transparency and smoothness of samples. J Adamovics is owner of Heuris Inc.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2017

Analysis of a Wave-Powered, Reverse-Osmosis System and Its Economic Availability in the United States: Preprint

A wave energy converter (WEC) system has the potential to convert the wave energy resource directly into the high-pressure flow that is needed by the desalination system to permeate saltwater through the reverse-osmosis membrane to generate clean water. In this study, a wave-to-water numerical model was developed to investigate the potential use of a wave-powered desalination system (WPDS) for water production in the United States. The model was developed by coupling a time-domain radiation-and-diffraction-method-based numerical tool (WEC-Sim) for predicting the hydrodynamic performance of WECs with a solution-diffusion model that was used to simulate the reverse-osmosis process. To evaluate the feasibility of the WPDS, the wave-to-water numerical model was applied to simulate a desalination system that used an oscillating surge WEC device to pump seawater through the system. The annual water production was estimated based on the wave resource at a reference site on the coast of northern California to investigate the potential cost of water in that area, where the cost of water and electricity is high compared to other regions. In the scenario evaluated, for a 100-unit utility-scale electricity-producing array, the estimated levelized cost of energy for these WECs is about 3-6 times the U.S.'s current, unsubsidized electricity rates. However, with clean water as an end product and by directly producing pressurized water with WECs, rather than electricity as an intermediary, it is presently only 12 percent greater than typical water cost in California. This study suggests that a WEC array that produces water may be a viable, near-term solution to the nation's water supply, and the niche application of the WPDS may also provide developers with new opportunities to further develop technologies that benefit both the electric and drinking water markets.

16 TIDAL AND WAVE POWER↗