Search NASA⌕ Search

SEARCH · Search NASA

Results for “Water Impact Testing”

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 253 records · Page 14

Investigation of electrochemical migration on Sn-0.7Cu-0.3Ag-0.03P-0.005Ni solder alloy in HNO{sub 3} solution

Current issue in lead-free solder in term of its reliability is still under investigation. This high impact research attempts to investigate the electrochemical migration (ECM) on Sn-0.7Cu-0.3Ag-0.03P-0.005Ni solder alloy by Water Drop Test (WDT) in different concentration of HNO{sub 3} solution. The concentration of HNO{sub 3} solution used in this research was 0.05, 0.10, 0.50 and 1M. Optical Microscope (OM), Field Emission Scanning Electron Microscope (FESEM) and Energy Dispersive X-Ray Analysis (EDX) were carried out in order to analysis the ECM behavior based on the growth of dendrite formation after WDT. In general, the results demonstrated that dendrite growth is faster in higher concentration compared with low concentration of HNO{sub 3}. The concentration of HNO{sub 3} solution used has a strong correlation with Mean-Time-To-Failure (MTTF). As the concentration of HNO{sub 3} increases, the MTTF value decreases. Based on the MTTF results the solder alloy in 1M HNO{sub 3} solution is most susceptible to ECM. SnO{sub 2} forms as a corrosion by-product in the samples proved by EDX analysis. The solder alloy poses a high reliability risk in microelectronic devices during operation in 1M HNO{sub 3} solution.

36 MATERIALS SCIENCE↗

Predicting Pit Stability On Additively Manufactured SS316 Via Finite Element Modeling [Slides]

Pit growth and repassivation are complex, with many interconnecting geometric and environmental parameters to consider. Experimentally, it is difficult to isolate these individual parameters to study their effect on the stability of pits. To enable these studies, a finite element modeling approach has been developed to allow systematic testing of parameters that impact a pit’s stability. The specific parameters studied were the cathode diameter, the pit diameter and shape, and the water layer thickness. Hemispherical and rectangular-based pits were studied to determine the impact of the overall pit shape. Pit stability results were compared with mathematical calculations based on the Maximum Pit Model, for both 50% saturation and 100% saturated salt film coverage. Further studies expanded the range of pit geometry to those relevant to additively manufactured surfaces.

36 MATERIALS SCIENCE↗

Characterization of Simulated BWR and Steam Exposed FeCrAl Coated Zircaloy 2

To combat the rapid oxidation that occurs in nuclear reactors during transient scenarios, investigations into modifying or replacing the existing zirconium based cladding and structural material have become a focus. Aside from testing these alternatives in the transient scenarios, verification of performance under normal operation is necessary. This work will examine the effect of normal BWR water chemistry over two time periods on FeCrAl coated Zircaloy 2 (Zry2) and its response to high temperature steam. X-ray diffraction (XRD), Scanning Electron Microscopy (SEM), Energy-dispersive X-ray Spectroscopy (EDS), and Thermogravimetric (TG) analysis are utilized to analyze the chemical and physical changes caused by these exposures. These tests show that under short time periods, normal water chemistry BWR conditions do not adversely change the effectiveness of a FeCrAl coating. A spinel structure was formed using Ni from an external source, however it did not have a major impact on the coatings performance in high temperature steam. Additional autoclave exposures up to 40 hours are ongoing, including coatings which were proton irradiated up to 5 dpa. (authors)

36 MATERIALS SCIENCE↗

Testing of Commercial Hollow Fiber Membranes for Space Suit Water Membrane Evaporator

Three commercial-off-the-shelf (COTS) hollow fiber (HoFi) membrane evaporators, modified for low pressure, were tested in a vacuum chamber at pressures below 33 pascals as potential space suit water membrane evaporator (SWME) heat rejection technologies. Water quality was controlled in a series of 25 tests, first simulating potable water reclaimed from waste water and then changing periodically to simulate the ever concentrating make-up of the circulating coolant over that is predicted over the course of 100 EVAs. Two of the systems, comprised of non-porous tubes with hydrophilic molecular channels as the water vapor transport mechanism, were severely impacted by the increasing concentrations of cations in the water. One of the systems, based on hydrophobic porous polypropylene tubes was not affected by the degrading water quality, or the presence of microbes. The polypropylene system, called SWME 1, was selected for further testing. An inverse flow configuration was also tested with SWME 1, with vacuum exposure on the inside of the tubes, provided only 20% of the performance of the standard configuration. SWME 1 was also modified to block 50% and 90% of the central tube layers, and tested to investigate performance efficiency. Performance curves were also developed in back-pressure regulation tests, and revealed important design considerations arising from the fully closed valve. SWME 1 was shown to be insensitive to air bubbles injected into the coolant loop. Development and testing of a full-scale prototype based on this technology and these test results is in progress.

Bue, Grant C.↗

SLUDGE BATCH 6/TANK 40 SIMULANT CHEMICAL PROCESS CELL SIMULATIONS

Phase III simulant flowsheet testing was completed using the latest composition estimates for SB6/Tank 40 feed to DWPF. The goals of the testing were to determine reasonable operating conditions and assumptions for the startup of SB6 processing in the DWPF. Testing covered the region from 102-159% of the current DWPF stoichiometric acid equation. Nitrite ion concentration was reduced to 90 mg/kg in the SRAT product of the lowest acid run. The 159% acid run reached 60% of the DWPF Sludge Receipt and Adjustment Tank (SRAT) limit of 0.65 lb H2/hr, and then sporadically exceeded the DWPF Slurry Mix Evaporator (SME) limit of 0.223 lb H2/hr. Hydrogen generation rates peaked at 112% of the SME limit, but higher than targeted wt% total solids levels may have been partially responsible for rates seen. A stoichiometric factor of 120% met both objectives. A processing window for SB6 exists from 102% to something close to 159% based on the simulant results. An initial recommendation for SB6 processing is at 115-120% of the current DWPF stoichiometric acid equation. The addition of simulated Actinide Removal Process (ARP) and Modular Caustic Side Solvent Extraction Unit (MCU) streams to the SRAT cycle had no apparent impact on the preferred stoichiometric factor. Hydrogen generation occurred continuously after acid addition in three of the four tests. The three runs at 120%, 118.4% with ARP/MCU, and 159% stoichiometry were all still producing around 0.1 lb hydrogen/hr at DWPF scale after 36 hours of boiling in the SRAT. The 120% acid run reached 23% of the SRAT limit and 37% of the SME limit. Conversely, nitrous oxide generation was subdued compared to previous sludge batches, staying below 29 lb/hr in all four tests or about a fourth as much as in comparable SB4 testing. Two processing issues, identified during SB6 Phase II flowsheet testing and qualification simulant testing, were monitored during Phase III. Mercury material balance closure was impacted by acid stoichiometry, and significant mercury was not accounted for in the highest acid run. Coalescence of elemental mercury droplets in the mercury water wash tank (MWWT) appeared to degrade with increasing stoichiometry. Observations were made of mercury scale formation in the SRAT condenser and MWWT. A tacky mercury amalgam with Rh, Pd, and Cu, plus some Ru and Ca formed on the impeller at 159% acid. It contained a significant fraction of the available Pd, Cu, and Rh as well as about 25% of the total mercury charged. Free (elemental) mercury was found in all of the SME products. Ammonia scrubbers were used during the tests to capture off-gas ammonia for material balance purposes. Significant ammonium ion formation was again observed during the SRAT cycle, and ammonia gas entered the off-gas as the pH rose during boiling. Ammonium ion production was lower than in the SB6 Phase II and the qualification simulant testing. Similar ammonium ion formation was seen in the ARP/MCU simulation as in the 120% flowsheet run. A slightly higher pH caused most of the ammonium to vaporize and collect in the ammonia scrubber reflux solution. Two periods of foaminess were noted. Neither required additional antifoam to control the foam growth. A steady foam layer formed during reflux in the 120% acid run. It was about an inch thick, but was 2-3 times more volume of bubbles than is typically seen during reflux. A similar foam layer also was seen during caustic boiling of the simulant during the ARP addition. While frequently seen with the radioactive sludge, foaminess during caustic boiling with simulants has been relatively rare. Two further flowsheet tests were performed and will be documented separately. One test was to evaluate the impact of process conditions that match current DWPF operation (lower rates). The second test was to evaluate the impact of SRAT/SME processing on the rheology of a modified Phase III simulant that had been made five times more viscous using ultrasonication.

36 MATERIALS SCIENCE↗

An exploration of the utilities of terahertz waves for the NDE of composites

We report an investigation of terahertz waves for the nondestructive evaluation of composite materials and structures. The modalities of the terahertz radiation used were time domain spectroscopy (TDS) and continuous wave (CW). The composite materials and structures investigated include both non-conducting polymeric composites and carbon fiber composites. Terahertz signals in the TDS mode resembles that of ultrasound; however, unlike ultrasound, a terahertz pulse can detect a crack hidden behind a larger crack. This was demonstrated in thick GFRP laminates containing double saw slots. In carbon composites the penetration of terahertz waves is quite limited and the detection of flaws is strongly affected by the angle between the electric field vector of the terahertz waves and the intervening fiber directions. The structures tested in this study include both solid laminates and honeycomb sandwiches. The defects and anomalies investigated by terahertz waves were foreign material inclusions, simulated disbond and delamination, mechanical impact damage, heat damage, and water or hydraulic fluid ingression. The effectiveness and limitations of terahertz radiation for the NDE of composites are discussed.

36 MATERIALS SCIENCE↗

Heat Pump Water Heater Technology Assessment Based on Laboratory Research and Energy Simulation Models: Preprint

This paper explores the laboratory performance of five integrated Heat Pump Water Heaters (HPWHs) across a wide range of operating conditions representative of US climate regions. Laboratory results demonstrate the efficiency of this technology under most of the conditions tested and show that differences in control schemes and design features impact the performance of the individual units. Theseresults were used to understand current model limitations, and then to bracket the energy savings potential for HPWH technology in various US climate regions. Simulation results show that HPWHs are expected to provide significant energy savings in many climate zones when compared to other types of water heaters (up to 64%, including impact on HVAC systems).

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Application of a Linear Input/Output Model to Tankless Water Heaters

In this study, the applicability of a linear input/output model to gas-fired, tankless water heaters has been evaluated. This simple model assumes that the relationship between input and output, averaged over both active draw and idle periods, is linear. This approach is being applied to boilers in other studies and offers the potential to make a small number of simple measurements to obtain the model parameters. These parameters can then be used to predict performance under complex load patterns. Both condensing and non-condensing water heaters have been tested under a very wide range of load conditions. It is shown that this approach can be used to reproduce performance metrics, such as the energy factor, and can be used to evaluate the impacts of alternative draw patterns and conditions.

15 GEOTHERMAL ENERGY↗

Multiple-time-stepping generalized hybrid Monte Carlo methods

Performance of the generalized shadow hybrid Monte Carlo (GSHMC) method [1], which proved to be superior in sampling efficiency over its predecessors [2–4], molecular dynamics and hybrid Monte Carlo, can be further improved by combining it with multi-time-stepping (MTS) and mollification of slow forces. We demonstrate that the comparatively simple modifications of the method not only lead to better performance of GSHMC itself but also allow for beating the best performed methods, which use the similar force splitting schemes. In addition we show that the same ideas can be successfully applied to the conventional generalized hybrid Monte Carlo method (GHMC). The resulting methods, MTS-GHMC and MTS-GSHMC, provide accurate reproduction of thermodynamic and dynamical properties, exact temperature control during simulation and computational robustness and efficiency. MTS-GHMC uses a generalized momentum update to achieve weak stochastic stabilization to the molecular dynamics (MD) integrator. MTS-GSHMC adds the use of a shadow (modified) Hamiltonian to filter the MD trajectories in the HMC scheme. We introduce a new shadow Hamiltonian formulation adapted to force-splitting methods. The use of such Hamiltonians improves the acceptance rate of trajectories and has a strong impact on the sampling efficiency of the method. Both methods were implemented in the open-source MD package ProtoMol and were tested on a water and a protein systems. Results were compared to those obtained using a Langevin Molly (LM) method [5] on the same systems. The test results demonstrate the superiority of the new methods over LM in terms of stability, accuracy and sampling efficiency. This suggests that putting the MTS approach in the framework of hybrid Monte Carlo and using the natural stochasticity offered by the generalized hybrid Monte Carlo lead to improving stability of MTS and allow for achieving larger step sizes in the simulation of complex systems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Prevention and management of silica scaling in membrane distillation using pH adjustment

Membrane scaling by silica is a major challenge in desalination, particularly for inland desalination of brackish groundwater or geothermal resources, which often contain high concentrations of silica and dissolved solids. Adjustment of feed pH may reduce silica scaling risk, which is important for inland facilities that operate at high water recoveries to reduce brine disposal costs. However, water recovery of reverse osmosis is also limited due to increased osmotic pressure with feed water concentration. Membrane distillation (MD) is a thermally driven membrane desalination technique that is not limited by increased osmotic pressure of the feed. In this investigation, pH adjustment was tested as a strategy to reduce silica scaling risk in the MD process. With feed water pH less than 5 or higher than 10, scaling impacts were negligible at silica concentrations up to 600 mg/L. Scaling rates were highest at neutral pH between 6 and 8. Cleaning strategies were also explored to remove silica scale from membranes. Cleaning using NaOH solutions at pH higher than 11 to induce dissolution of silica scale was effective at temporarily restoring performance; however, some silica remained on membrane surfaces and scaling upon re-exposure to supersaturated silica concentrations occurred faster than with new membranes.

15 GEOTHERMAL ENERGY↗

Interactions Between Thresholds and Spatial Discretizations of Snow: Insights From Estimates of Wolverine Denning Habitat in the Colorado Rocky Mountains

Thresholds can be used to interpret environmental data in a way that is easily communicated and useful for decision making purposes. However, thresholds are often developed for specific data products and time periods, changing findings when the same threshold is applied to datasets or periods with different characteristics. Here, we test the impact of different spatial discretizations of snow on annual estimates of wolverine denning opportunities in the Colorado Rocky Mountains, defined using a snow water equivalent (SWE) threshold (0.20 m) and threshold date (15 May) from previous habitat assessments. Annual potential wolverine denning area (PWDA) was thresholded from a 36-year (1985 – 2020) snow reanalysis model with three different spatial discretizations: 1) 480 m grid cells (D480), 2) 90 m grid cells (D90), and 3) 480 m grid cells with implicit representations of subgrid snow spatial heterogeneity (S480). Relative to the D480 and S480 discretizations, D90 resolved shallower snow deposits on slopes between 3050 and 3350 m elevation, decreasing PWDA by 10%, on average. In years with warmer and/or drier winters, S480 discretizations with subgrid representations of snow heterogeneity increased PWDA, even within grid cells where mean 15 May SWE was less than the SWE threshold. These simulations increased PWDA by upwards of 30% in low snow years, as compared to the D480 and D90 simulations without subgrid snow heterogeneity. Despite PWDA sensitivity to different snow spatial discretizations, PWDA was controlled more by annual variations in winter precipitation and temperature. However, small changes to the SWE threshold (± 0.07 m) and threshold date (± 2 weeks) also affected PWDA by as much as 82%. Across these threshold ranges, PWDA was approximately 18% more sensitive to the SWE threshold than the threshold date. However, the sensitivity to the threshold date was larger in years with late spring snowfall, when PWDA depended on whether modeled SWE was thresholded before, during, or after spring snow accumulation. Our results demonstrate that snow thresholds are useful but may not always provide a complete picture of the annual variability in snow-adapted wildlife denning opportunities. Studies thresholding spatiotemporal datasets could be improved by including 1) information about the fidelity of thresholds across multiple spatial discretizations, and 2) uncertainties related to ranges of realistic thresholds.

Justin M. Pflug↗

Parametric Study of SANSMIC Input and Resulting Impact on Predicted Cavern Geometry and Leaching Efficiency

The Sandia Solution Mining Code (SANSMIC) has been used for many years to examine the development of salt cavern geometry, both in a confirmatory manner with comparisons made to real-world sonar data and in a predictive manner when updated sonar data are not available. SANSMIC models require some modeling choices in order to incorporate real-world data. Key modeling choices include the vertical resolution of cavern geometry to implement, as well as how to incorporate daily raw water injection data into the SANSMIC model. This report documents five studies that address the impact of the modeling choices on the predicted cavern geometries and calculated leaching efficiencies. In most cases, hypothetical cylindrical initial cavern geometries are used to provide a common baseline against which to test the systematic variation of input variables including cavern radius, oil-brine-interface (OBI) depth, vertical cell size, raw water injection rate, raw water injection duration, workover time, and number of leaching stages. The use of smaller cell sizes is recommended moving forward to provide a better one-to-one relationship between sonar data and the modeled cavern. A new methodology for incorporating raw water injection data is also recommended, in order to more closely model real-world injection and workover times. Overall, the systematic studies performed here have increased our confidence in previous SANSMIC model results, as well future use of the code for predicting leaching effects on cavern geometries. Some minor changes to modeling choices are recommended, which can easily be applied with the version of SANSMIC currently under development.

42 ENGINEERING↗

Linking Ceragenins to Water-Treatment Membranes to Minimize Biofouling

Ceragenins were used to create biofouling resistant water-treatment membranes. Ceragenins are synthetically produced antimicrobial peptide mimics that display broad-spectrum bactericidal activity. While ceragenins have been used on bio-medical devices, use of ceragenins on water-treatment membranes is novel. Biofouling impacts membrane separation processes for many industrial applications such as desalination, waste-water treatment, oil and gas extraction, and power generation. Biofouling results in a loss of permeate flux and increase in energy use. Creation of biofouling resistant membranes will assist in creation of clean water with lower energy usage and energy with lower water usage. Five methods of attaching three different ceragenin molecules were conducted and tested. Biofouling reduction was observed in the majority of the tests, indicating the ceragenins are a viable solution to biofouling on water treatment membranes. Silane direct attachment appears to be the most promising attachment method if a high concentration of CSA-121a is used. Additional refinement of the attachment methods are needed in order to achieve our goal of several log-reduction in biofilm cell density without impacting the membrane flux. Concurrently, biofilm forming bacteria were isolated from source waters relevant for water treatment: wastewater, agricultural drainage, river water, seawater, and brackish groundwater. These isolates can be used for future testing of methods to control biofouling. Once isolated, the ability of the isolates to grow biofilms was tested with high-throughput multiwell methods. Based on these tests, the following species were selected for further testing in tube reactors and CDC reactors: Pseudomonas ssp. (wastewater, agricultural drainage, and Colorado River water), Nocardia coeliaca or Rhodococcus spp. (wastewater), Pseudomonas fluorescens and Hydrogenophaga palleronii (agricultural drainage), Sulfitobacter donghicola, Rhodococcus fascians, Rhodobacter katedanii, and Paracoccus marcusii (seawater), and Sphingopyxis spp. (groundwater). The testing demonstrated the ability of these isolates to be used for biofouling control testing under laboratory conditions. Biofilm forming bacteria were obtained from all the source water samples.

59 BASIC BIOLOGICAL SCIENCES↗

2011 Annual Industrial Wastewater Reuse Report for the Idaho National Laboratory Site's Advanced Test Reactor Complex Cold Waste Pond

This report describes conditions, as required by the state of Idaho Industrial Wastewater Reuse Permit (LA 000161 01, Modification B), for the wastewater land application site at the Idaho National Laboratory Site's Advanced Test Reactor Complex Cold Waste Pond from November 1, 2010 through October 31, 2011. The report contains the following information: Facility and system description Permit required effluent monitoring data and loading rates Groundwater monitoring data Status of compliance activities Noncompliance and other issues Discussion of the facility's environmental impacts During the 2011 permit year, approximately 166 million gallons of wastewater were discharged to the Cold Waste Pond. This is well below the maximum annual permit limit of 375 million gallons. As shown by the groundwater sampling data, sulfate and total dissolved solids concentrations are highest near the Cold Waste Pond and decrease rapidly as the distance from the Cold Waste Pond increases. Although concentrations of sulfate and total dissolved solids are elevated near the Cold Waste Pond, both parameters were below the Ground Water Quality Rule Secondary Constituent Standards in the down gradient monitoring wells.

2011 Annual Industrial Wastewater Reuse Report↗

Mineral Scaling in 3D Interfacial Solar Evaporators–A Challenge for Brine Treatment and Lithium Recovery

In this work, we analyzed the effects of mineral scaling on the performance of a 3D interfacial solar evaporator, with a focus on the cations relevant to lithium recovery from brackish water. The field has been rapidly moving toward resource recovery applications from brines with higher cation concentrations. However, the potential complications caused by common minerals in these brines other than NaCl have been largely overlooked. Therefore, in this study, we thoroughly examined the effects of two common cations (calcium and magnesium) on the long-term solar evaporation performance of a 3D graphene oxide stalk. The 3D stalk can achieve an evaporation flux as high as 17.8 kg m –2 h –1 under one-sun illumination, and accumulation of NaCl on the stalk surface has no impact. However, the presence of CaCl 2 and MgCl 2 significantly reduces the evaporative flux even in solutions lacking scale-forming anions. A close examination of scale formation during long-term evaporation experiments revealed that CaCl 2 and MgCl 2 tend to precipitate out within the stalk, thus blocking water transport through the stalk and significantly dropping the evaporation rates. These findings imply that research attention is needed to modify and optimize the internal water transport channels for 3D evaporators. Additionally, we emphasize the importance of testing realistic mixtures–including prominent divalent cations– and testing long-term operations in interfacial solar evaporation research and investigating approaches to mitigate the negative impacts of divalent cations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Testing, Modeling and System Impact of Metabolic Heat Regenerated Temperature Swing Adsorption

Metabolic heat regenerated temperature swing adsorption (MTSA) technology is being developed for removal and rejection of carbon dioxide (CO2) and heat from a portable life support system (PLSS) to the Martian environment. Previously, hardware was built and tested to demonstrate using heat from simulated, dry ventilation loop gas to affect the temperature swing required to regenerate an adsorbent used for CO2 removal. New testing has been performed using a moist, simulated ventilation loop gas to demonstrate the effects of water condensing and freezing in the heat exchanger during adsorbent regeneration. In addition, thermal models of the adsorbent during regeneration were modified and calibrated with test data to capture the effect of the CO2 heat of desorption. Finally, MTSA impact on PLSS design was evaluated by performing thermal balances assuming a specific PLSS architecture. Results using NASA s Extravehicular Activity System Sizing Analysis Tool (EVAS_SAT), a PLSS system evaluation tool, are presented.

Lacomini, Christine S.↗

Two-moment Bulk Stratiform Cloud Microphysics in the Grid-point Atmospheric Model of IAP LASG (GAMIL)

A two-moment bulk stratiform microphysics scheme, including recently developed physically-based droplet activation/ice nucleation parameterizations has been implemented into the Grid-point Atmospheric Model of IAP LASG (GAMIL) as an effort to enhance the model capability for studying aerosol indirect effects. Unlike the previous one-moment cloud microphysics scheme, the new scheme produces reasonable representation of cloud particle size and number concentration. This scheme captures the observed spatial variations in cloud droplet number concentrations. Simulated ice crystal number concentrations in cirrus clouds qualitatively agree with in-situ observations. The longwave and shortwave cloud forcing are in better agreement with observations. Sensitivity tests show that the column cloud droplet number concentrations calculated from two different droplet activation parameterizations are similar. However, ice crystal number concentration in mixed-phased clouds is sensitive to different heterogeneous freezing formulations. The simulation with high ice crystal number concentration in mixed-phase clouds has less liquid water path and weaker cloud forcing. Furthermore, ice crystal number concentration in cirrus clouds is sensitive to different ice nucleation parameterizations. Sensitivity tests also suggest that impact of pre-existing ice crystals on homogeneous freezing in old clouds should be taken into account.

(GAMIL)↗

An investigation on machine learning predictive accuracy improvement and uncertainty reduction using VAE-based data augmentation

The confluence of ultrafast computers with large memory, rapid progress in Machine Learning (ML) algorithms, and the availability of large datasets place multiple engineering fields at the threshold of dramatic progress. However, a unique challenge in nuclear engineering is data scarcity because experimentation on nuclear systems is usually more expensive and time-consuming than most other disciplines. One potential way to resolve the data scarcity issue is deep generative learning, which uses certain ML models to learn the underlying distribution of existing data and generate synthetic samples that resemble the real data. In this way, one can significantly expand the dataset to train more accurate predictive ML models. In this study, our objective is to evaluate the effectiveness of data augmentation using variational autoencoder (VAE)-based deep generative models. We investigated whether the data augmentation leads to improved accuracy in the predictions of a deep neural network (DNN) model trained using the augmented data. Additionally, the DNN prediction uncertainties are quantified using Bayesian Neural Networks (BNN) and conformal prediction (CP) to assess the impact on predictive uncertainty reduction. To test the proposed methodology, we used TRACE simulations of steady-state void fraction data based on the NUPEC Boiling Water Reactor Full-size Fine-mesh Bundle Test (BFBT) benchmark. Here, we found that augmenting the training dataset using VAEs has improved the DNN model’s predictive accuracy, improved the prediction confidence intervals, and reduced the prediction uncertainties.

Bayesian neural network↗