Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Predictive Maintenance”

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 847 records · Page 47

ITER's Tokamak Cooling Water System and the the Use of ASME Codes to Comply with French Regulations of Nuclear Pressure Equipment

During inductive plasma operation of ITER, fusion power will reach 500 MW with an energy multiplication factor of 10. The heat will be transferred by the Tokamak Cooling Water System (TCWS) to the environment using the secondary cooling system. Plasma operations are inherently safe even under the most severe postulated accident condition a large, in-vessel break that results in a loss-of-coolant accident. A functioning cooling water system is not required to ensure safe shutdown. Even though ITER is inherently safe, TCWS equipment (e.g., heat exchangers, piping, pressurizers) are classified as safety important components. This is because the water is predicted to contain low-levels of radionuclides (e.g., activated corrosion products, tritium) with activity levels high enough to require the design of components to be in accordance with French regulations for nuclear pressure equipment, i.e., the French Order dated 12 December 2005 (ESPN). ESPN has extended the practical application of the methodology established by the Pressure Equipment Directive (97/23/EC) to nuclear pressure equipment, under French Decree 99-1046 dated 13 December 1999, and Order dated 21 December 1999 (ESP). ASME codes and supplementary analyses (e.g., Failure Modes and Effects Analysis) will be used to demonstrate that the TCWS equipment meets these essential safety requirements. TCWS is being designed to provide not only cooling, with a capacity of approximately 1 GW energy removal, but also elevated temperature baking of first-wall/blanket, vacuum vessel, and divertor. Additional TCWS functions include chemical control of water, draining and drying for maintenance, and facilitation of leak detection/localization. The TCWS interfaces with the majority of ITER systems, including the secondary cooling system. U.S. ITER is responsible for design, engineering, and procurement of the TCWS with industry support from an Engineering Services Organization (ESO) (AREVA Federal Services, with support from Northrop Grumman, and OneCIS). ITER International Organization (ITER-IO) is responsible for design oversight and equipment installation in Cadarache, France. TCWS equipment will be fabricated using ASME design codes with quality assurance and oversight by an Agreed Notified Body (approved by the French regulator) that will ensure regulatory compliance. This paper describes the TCWS design and how U.S. ITER and fabricators will use ASME codes to comply with EU Directives and French Orders and Decrees.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Development of Lightweight Structural Materials with Improved Properties for Fission Batteries

The notion of a “fission battery” conveys a vision focused on realizing very simple “plug-and-play” nuclear systems that can be integrated into a variety of applications requiring affordable, reliable energy in the form of electricity and/or heat and function without operations and maintenance staff. Fission batteries require lightweight structural materials to increase their mobility, and the lightweight materials must demonstrate structural resilience under various conditions. The objective of this work is to develop lightweight structural material featuring a good combination of mechanical properties using advanced modeling and simulation together with an advanced additive manufacturing technique such as laser powder bed fusion. The preliminary results show that different lattice structures and types can be successfully meshed using nTopology software, and the lattice structure data can be successfully transformed to Multiphysics Object-Oriented Simulation Environment (MOOSE) input. Finite Element Analysis (FEA) displays that, at macro/engineering scale simulation, the weight saving design has an obvious effect on tensile behavior such as effective elastic modulus and yield stress. The novel approaches of this work are (1) development of optimized lattice structures for improved mechanical properties using advanced simulation and modeling techniques; and (2) model predictions of the mechanical properties (e.g., strength and stress distribution) of macroscopic materials in order to preliminarily select a lattice structure for additive manufacturing.

36 MATERIALS SCIENCE↗

Volumetrically Absorbing Thermal Insulator (VATI) for High-temperature Receivers

This project seeks to exploit volumetric absorption of concentrated solar irradiation in a thin, high-temperature, thermally conductive medium. High efficiency thermal conversion is achieved by volumetric absorption of both incoming solar irradiation (short wavelength) and emitted irradiation (long wavelength), demonstrating a volumetrically absorbing thermally insulating (VATI) effect. The thermalized energy is conducted to the back wall, enclosing the working fluid (molten salt or supercritical CO 2 ). Reliance on conductive thermal transport requires a thermally conductive medium, in the absence of which large temperature gradients drive losses due to emission. The project’s outcomes are important to realize a cost-effective approach to reduce optical and thermal losses from CSP receivers at high temperatures (720°C). High-temperature stable and commercially available porous SiC structures (open-cell foams) were explored as a volumetrically absorbing and thermally insulating layer. To the best of our knowledge, the current state-of-the-art for industrially deployed coatings is Pyromark. However, Pyromark suffers degradation at 700+°C and diurnal temperature changes. This necessitates periodic recoating, and the downtime increases the overall levelized cost of energy (LCOE) production. On the other hand, contemporary research activities have generated significant advances in the development of selective emitter coatings, which present challenges with costs, scalability, and stability. The pursued approach alleviates these concerns by developing a receiver that utilizes the inherent structure of high-temperature stable porous materials to enable robust and cost-effective receivers which require no periodic maintenance downtimes. The overall goal of the project is to experimentally demonstrate a figure of merit (FOM) of 0.92 at a temperature of 720°C and solar irradiation of 1000x concentration with porous receivers. The relevant crystallographic (phase) optical and thermal properties of porous SiC were first characterized. Second, the 3-D geometry of the porous SiC was analyzed using micro-X-ray tomography and converted to CAD data using image processing analysis. Utilizing this 3-D geometry and relevant optical/thermal properties, Monte Carlo- Ray Tracing (MCRT) analysis was performed to extract important parameters governing solar-thermal energy conversion such as extinction coefficient (β, 1/m), scattering albedo (ω) and the scattering phase function (Φ). These properties were then integrated into an in-house radiative and conduction transport model to solve for temperature and transport fluxes characterizing the solar-thermal energy conversion. This model was utilized to predict the FOM for various SiC porous geometries and to identify the highest possible FOM. Testing of the optimized porous structures will be accomplished with a custom-built high-accuracy (< ±4%) FOM measurement test-stand and a 1000x solar concentrator. When neglecting convective losses and resistance at the open boundary and the back wall, the optimized SiC foam leads to a FOM of 0.84. This FOM does not surpass the performance of Pyromark 2500. Yet, conversely to Pyromark 2500, SiC is stable at high temperatures and does not degrade over time. Also, it is possible to boost the FOM of SiC by engineering its effective thermal conductivity and its scattering albedo. A FOM of 0.92 is predicted for an optimized foam by accounting for convective losses and resistance at the open boundary and the back wall. This largely exceeds the FOM of Pyromark 2500 (~0.86) predicted by neglecting convective losses and resistance. Therefore, further engineering of the foam could lead to unprecedented FOMs.

14 SOLAR ENERGY↗

Development of Lightweight Structural Materials with Improved Properties for Fission Batteries

The notion of a “fission battery” conveys a vision focused on realizing very simple “plug-and-play” nuclear systems that can be integrated into a variety of applications requiring affordable, reliable energy in the form of electricity and/or heat and function without operations and maintenance staff. Fission batteries require lightweight structural materials to increase their mobility, and the lightweight materials must demonstrate structural resilience under various conditions. The objective of this work is to develop lattice structured lightweight structural material featuring a good combination of mechanical properties using advanced modeling and simulation together with an advanced additive manufacturing technique such as laser powder bed fusion. The preliminary results show that different lattice structures and types can be successfully meshed using nTopology software, and the lattice structure data can be successfully transformed to Multiphysics Object-Oriented Simulation (MOOSE) Environment input. Finite Element Analysis (FEA) displays that, at macro/engineering scale simulation, the weight saving design has an obvious effect on tensile behavior such as effective elastic modulus and yield stress. The novel approaches of this work are (1) development of lattice structures for improved mechanical properties using advanced simulation and modeling techniques; and (2) model predictions of the mechanical properties (e.g., strength and stress distribution) of macroscopic materials in order to preliminarily select a lattice structure for additive manufacturing.

36 MATERIALS SCIENCE↗

COMPUTER-AIDED LATTICE DESIGN AND ADVANCED MODELING FOR THE DEVELOPMENT OF LIGHTWEIGHT STRUCTURAL MATERIALS

The notion of a “fission battery” conveys a vision focused on realizing very simple “plug-and-play” nuclear systems that can be integrated into a variety of applications requiring affordable, reliable energy in the form of electricity and/or heat and function without operations and maintenance staff. Fission batteries require lightweight structural materials to increase their mobility, and the lightweight materials must demonstrate structural resilience under various conditions. The objective of this work is to develop lattice structured lightweight structural material featuring a good combination of mechanical properties using advanced modeling and simulation together with an advanced additive manufacturing technique such as laser powder bed fusion. The preliminary results show that different lattice structures and types can be successfully meshed using nTopology software, and the lattice structure data can be successfully transformed to Multiphysics Object-Oriented Simulation (MOOSE) Environment input. Finite Element Analysis (FEA) displays that, at macro/engineering scale simulation, the weight saving design has an obvious effect on tensile behavior such as effective elastic modulus and yield stress. The novel approaches of this work are (1) development of lattice structures for improved mechanical properties using advanced simulation and modeling techniques; and (2) model predictions of the mechanical properties (e.g., strength and stress distribution) of macroscopic materials in order to preliminarily select a lattice structure for additive manufacturing.

36 MATERIALS SCIENCE↗

Characterization and Comparison of Vibration Transfer Paths in a Helicopter Gearbox and a Fixture Mounted Gearbox

Health monitoring of rotorcraft components, currently being performed by Health and Usage Monitoring Systems through analyses of vibration signatures of dynamic mechanical components, is very important for their safe and economic operation. HUMS analyze vibration signatures associated with faults and quantify them as condition indicators to predict component behavior. Vibration transfer paths are characterized by frequency response functions derived from the input/output relationship between applied force and dynamic response through a structure as a function of frequency. With an objective to investigate the differences in transfer paths, transfer path measurements were recorded under similar conditions in the left and right nose gearboxes of an AH-64 helicopter and in an isolated left nose gearbox in a test fixture at NASA Glenn Research Center. The test fixture enabled the application of measured torques-common during an actual operation. An impact hammer as well as commercial and lab piezo shakers, were used in conjunction with two types of commercially available accelerometers to collect the vibration response under various test conditions. The frequency response functions measured under comparable conditions of both systems were found to be consistent. Measurements made on the fixture indicated certain real-world installation and maintenance issues, such as sensor alignments, accelerometer locations and installation torques, had minimal effect. However, gear vibration transfer path dynamics appeared to be somewhat dependent on the presence of oil, and the transfer path dynamics were notably different if the force input was on the internal ring gear rather than on the external gearbox case.

structures↗

Lunar Limb Observatory: An Incremental Plan for the Utilization, Exploration, and Settlement of the Moon

This paper proposes a comprehensive incremental program, Lunar Limb Observatory (LLO), for a return to the Moon, beginning with robotic missions and ending with a permanent lunar settlement. Several recent technological developments make such a program both affordable and scientifically valuable: robotic telescopes, the Internet, light-weight telescopes, shared- autonomy/predictive graphics telerobotic devices, and optical interferometry systems. Reasons for focussing new NASA programs on the Moon include public interest, Moon-based astronomy, renewed lunar exploration, lunar resources (especially helium-3), technological stimulus, accessibility of the Moon (compared to any planet), and dispersal of the human species to counter predictable natural catastrophes, asteroidal or cometary impacts in particular. The proposed Lunar Limb Observatory would be located in the crater Riccioli, with auxiliary robotic telescopes in M. Smythii and at the North and South Poles. The first phase of the program, after site certification, would be a series of 5 Delta-launched telerobotic missions to Riccioli (or Grimaldi if Riccioli proves unsuitable), emplacing robotic telescopes and carrying out surface exploration. The next phase would be 7 Delta-launched telerobotic missions to M. Smythii (2 missions), the South Pole (3 missions), and the North Pole (2 missions), emplacing robotic telescopes to provide continuous all-sky coverage. Lunar base establishment would begin with two unmanned Shuttle/Fitan-Centaur missions to Riccioli, for shelter emplacement, followed by the first manned return, also using the Shuttle/Fitan-Centaur mode. The main LLO at Riccioli would then be permanently or periodically inhabited, for surface exploration, telerobotic rover and telescope operation and maintenance, and support of Earth-based student projects. The LLO would evolve into a permanent human settlement, serving, among other functions, as a test area and staging base for the exploration, settlement, and terraforming of Mars.

Lowman, Paul. D., Jr.↗

Sampling for Air Chemical Emissions from the Life Sciences Laboratory II

Sampling for air chemical emissions from the Life Science Laboratory II (LSL-II) ventilation stack was performed in an effort to determine potential exposure of maintenance staff to laboratory exhaust on the building roof. The concern about worker exposure was raised in December 2015 and several activities were performed to assist in estimating exposure concentrations. Data quality objectives were developed to determine the need for and scope and parameters of a sampling campaign to measure chemical emissions from research and development activities to the outside air. The activities provided data on temporal variation of air chemical concentrations and a basis for evaluating calculated emissions. Sampling for air chemical emissions was performed in the LSL-II ventilation stack over the 6-week period from July 26 to September 1, 2016. A total of 12 sampling events were carried out using 16 sample media. Resulting analysis provided concentration data on 49 analytes. All results were below occupational exposure limits and most results were below detection limits. When compared to calculated emissions, only 5 of the 49 chemicals had measured concentrations greater than predicted. This sampling effort will inform other study components to develop a more complete picture of a worker’s potential exposure from LSL-II rooftop activities. Mixing studies were conducted to inform spatial variation in concentrations at other rooftop locations and can be used in conjunction with these results to provide temporal variations in concentrations for estimating the potential exposure to workers working in and around the LSL-II stack.

54 ENVIRONMENTAL SCIENCES↗

Determination of a Generic Human Error Probability Distribution, Part 2: A Dynamic SPAR-H Example Application

Humans are an integral part of a nuclear power plant (NPP) from its construction, to its everyday maintenance and operation. In the nuclear industry there are many approaches to quantifying component reliability, which is called probability risk assessment (PRA). In PRA there are two considerations given to the quantification, static and dynamic. Static is quantified before or after an operation; and dynamic is time dependent, in that the probability changes as real time data are provided. In PRA, the importance of the components is calculated, but what of the importance of the human component? There are several methods for human reliability analysis (HRA), which include but are not limited to: THERP, ATHEANA, CREAM, SPAR-H, ASEP, and SLIM. Each of these methods aims to provide a more accurate and objective human error probability (HEP) quantification. Of the methods listed above, ASEP, THERP, and SPAR-H are all inextricably linked. The Standardized Plant Analysis Risk-Human Reliability Analysis (SPAR-H) method was generated to better generalize the Technique for Human Error Rate Prediction (THERP) and Accident Sequence Precursor (ASEP). THERP is mapped to specific scenario templates, and ASEP is a simplified screening method, while SPAR-H employs generalizations to describe the spectrum of human behavior while maintaining the same theoretical underpinnings. SPAR-H is a well-known and accepted HRA method that uses performance shaping factors (PSFs) to classify inputs to the human component. PSFs capture a wide variety of input data, including plant status, crew dynamics, task description, and psychological aspects of the human operator. Current efforts have been focused to further generalize SPAR-H in to a dynamic framework. These methods can fit into dynamic NPP simulations such as RAVEN / RELAP-7. In order to do this, a dynamic methodology is applied to model the human as would be applied to the components. This method is applied in part one in the accompanying paper.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Managing an Aging Pump-and-Treat Groundwater Remediation System at the Fernald Preserve - 19557

The Fernald Preserve, Ohio, Site underwent restoration and remediation under the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). With the exception of the Great Miami Aquifer, the CERCLA remediation was declared physically complete on October 29, 2006, and the site was turned over to the US Department of Energy Office of Legacy Management for long-term care and maintenance. The Record of Decision under CERCLA specifies a concentration-based cleanup that requires the extraction and treatment of Great Miami Aquifer groundwater above Final Remediation Levels (FRLs) until the full, beneficial use potential of the aquifer is achieved, including use as a drinking water source. The area of the aquifer targeted for remediation is 126.5 hectares (312.70 acres). The groundwater pumping remediation was designed by modeling cleanup with the Variable Saturated Analysis Model in 3 Dimensions (VAM-3D) modeling code, developed by Hydrogeologic Inc. The original design called for 23 extraction wells operating at 379 liters per minute (lpm) (100 gallons per minute [gpm]) to 1135 lpm (300 gpm) each for a system total of 18,073 lpm (4775 gpm). In 2014, the system was optimized down to 20 extraction wells operating at 379 to 1893 lpm each for a system total of 19,180 lpm (5075 gpm). To date, pumping has been an effective remediation technology for the aquifer. At the end of 2017, the area of the aquifer remaining above FRLs was down to 38.2 hectares (94.4 acres). From August 1993 through December 2017 (24 years), 6235 kilograms (kg) (13,733 pounds) of uranium were removed from the aquifer by the pumping of 1.74 x 10{sup 11} liters (46.2 billion gallons) of water. Since 2010, treatment of groundwater to achieve regulatory discharge limits at the Great Miami River has not been needed on a routine basis. The ongoing remediation is now essentially a pumping operation, and this water does not require treatment. Metrics reported in the annual Site Environmental Report for the groundwater remediation indicate that, based on the model-predicted estimate of the mass of uranium that needs to be removed from the aquifer to achieve a concentration-based cleanup, the cleanup is approximately 86% - 87% complete. Current model predictions indicate that pumping will need to continue an additional 16 years (until 2035) in some of the wells. During the additional 16 years, it is estimated that approximately 1012 kg (2231 pounds) of uranium will be removed from the aquifer. System pumping rates are designed to decrease over the course of the remaining 16 years, dropping down to a system total of 11,638 lpm (3075 gpm) in 2022 and a system total of 4163 lpm (1100 gpm) in 2033. Even with the designed decreases in pumping rates, it is estimated that approximately 1 x 10{sup 11} liters (26.6 billion gallons) of water will be pumped from the aquifer to remove 1012 kg (2231 pounds) of uranium. This indicates that the remaining 16 years of operation will be much less efficient than the first 24 years. Decreasing efficiency is common for pumping operations. The worst-case scenario is that decreasing concentration trends stabilize above the cleanup FRLs and remediation goals are never achieved. Performance assessments of the pumping remedy will need to be routinely conducted to determine if pumping should be discontinued, if it should be continued but optimized, or if the pumping technology can be supplemented with a different technology that will improve the outcome of the remediation and make the operation more efficient and successful. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

Overlap Properties of Clouds Generated by a Cloud Resolving Model

In order for General Circulation Models (GCMs), one of our most important tools to predict future climate, to correctly describe the propagation of solar and thermal radiation through the cloudy atmosphere a realistic description of the vertical distribution of cloud amount is needed. Actually, one needs not only the cloud amounts at different levels of the atmosphere, but also how these cloud amounts are related, in other words, how they overlap. Currently GCMs make some idealized assumptions about cloud overlap, for example that contiguous cloud layers overlap maximally and non-contiguous cloud layers overlap in a random fashion. Since there are difficulties in obtaining the vertical profile of cloud amount from observations, the realism of the overlap assumptions made in GCMs has not been yet rigorously investigated. Recently however, cloud observations from a relatively new type of ground radar have been used to examine the vertical distribution of cloudiness. These observations suggest that the GCM overlap assumptions are dubious. Our study uses cloud fields from sophisticated models dedicated to simulate cloud formation, maintenance, and dissipation called Cloud Resolving Models . These models are generally considered capable of producing realistic three-dimensional representation of cloudiness. Using numerous cloud fields produced by such a CRM we show that the degree of overlap between cloud layers is a function of their separation distance, and is in general described by a combination of the maximum and random overlap assumption, with random overlap dominating as separation distances increase. We show that it is possible to parameterize this behavior in a way that can eventually be incorporated in GCMs. Our results seem to have a significant resemblance to the results from the radar observations despite the completely different nature of the datasets. This consistency is encouraging and will promote development of new radiative transfer codes that will estimate the radiation effects of multi-layer cloud fields more accurately.

Oreopoulos, L.↗

Modeling of Vertical Motor-driven Pump for Simulation of a Fault Signature \\ for Condition Monitoring

As part of the ongoing effort to transition from preventive maintenance strategies to condition-based maintenance strategies in nuclear power plants, there is significant reliance on using machine learning techniques. To develop a robust machine learning model that can diagnose all the fault modes of a vertical motor-driven pump, data capturing the unique signature of each fault mode is required. In practice, it is difficult to collect or capture data that captures all the fault modes from a single plant site. So to address this situation, a computational model of a vertical motor-driven pump is developed using the multipurpose finite element software COMSOL Multiphysics. The developed model is used to generate simulated data under normal operation and is compared with the vibration data collected using vibration sensors. Once the simulation model is verified under normal operating condition, simulated data for the fault mode for which minimal or no evidence is available in historical plant process data is developed. This simulated data is used to develop fault signatures to achieve robust predictive models. This paper presents modeling details and verification of the model that can used to generate data for fault modes that are not available at a plant site for condition monitoring purpose.

46 - INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AN↗

Modeling of Vertical Motor-driven Pump for Simulation of a Fault Signature \\ for Condition Monitoring

As part of the ongoing effort to transition from preventive maintenance strategies to condition-based maintenance strategies in nuclear power plants, there is significant reliance on using machine learning techniques. To develop a robust machine learning model that can diagnose all the fault modes of a vertical motor-driven pump, data capturing the unique signature of each fault mode is required. In practice, it is difficult to collect or capture data that captures all the fault modes from a single plant site. So to address this situation, a computational model of a vertical motor-driven pump is developed using the multipurpose finite element software COMSOL Multiphysics. The developed model is used to generate simulated data under normal operation and is compared with the vibration data collected using vibration sensors. Once the simulation model is verified under normal operating condition, simulated data for the fault mode for which minimal or no evidence is available in historical plant process data is developed. This simulated data is used to develop fault signatures to achieve robust predictive models. This paper presents modeling details and verification of the model that can used to generate data for fault modes that are not available at a plant site for condition monitoring purpose.

46 - INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AN↗

Sensor impact evaluation and verification for fault detection and diagnostics in building energy systems: A review

Sensors are the key information source for fault detection and diagnostics (FDD) in buildings. However, sensors are often not properly designed, installed, calibrated, located, and maintained, which negatively impacts FDD performance. Several sensor-related FDD topics have been widely studied, covering a wide range of fault types and applications. However, it is difficult to get a clear picture of the technical development of sensor-related topics in FDD. A systematic review of sensor topics is needed to summarize the existing research in a logical way, draw conclusions on the current development, and predict the future development of sensors in building FDD. To address this gap, we conducted a comprehensive literature review of more than 100 FDD-sensor-related papers. In this article, we subdivide the FDD tasks into building-level, system-level, and component-level FDD, and review sensor-related topics in each category. Our major conclusions are: (a) current data-driven FDD research focuses more on FDD algorithms than sensors, (b) sensor “hardware” research topics are less studied than sensor “software” topics, (c) very few papers focus on sensor engineering as an integral aspect of FDD development, and (d) some important sensor topics, such as sensor cost-effectiveness and sensor schema/layout/location, are not well studied. Finally, we discuss the need for a systematic framework of FDD sensors and models to integrate sensor design/selection, sensor data analysis/mining, feature selection, physics-based or data-driven algorithm development, sensor fault detection, sensor calibration, and sensor maintenance. Finally, expert interviews are conducted to validate the above findings and conclusions.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Transformers and Long Short-Term Memory Transfer Learning for GenIV Reactor Temperature Time Series Forecasting

Automated monitoring of the coolant temperature can enable autonomous operation of generation IV reactors (GenIV), thus reducing their operating and maintenance costs. Automation can be accomplished with machine learning (ML) models trained on historical sensor data. However, the performance of ML usually depends on the availability of large amount of training data, which is difficult to obtain for GenIV, as this technology is still under development. We propose the use of transfer learning (TL), which involves utilizing knowledge across different domains, to compensate for this lack of training data. TL can be used to create pre-trained ML models with data from small-scale research facilities, which can then be fine-tuned to monitor GenIV reactors. In this work, we develop pre-trained Transformer and long short-term memory (LSTM) networks by training them on temperature measurements from thermal hydraulic flow loops operating with water and Galinstan fluids at room temperature at Argonne National Laboratory. The pre-trained models are then fine-tuned and re-trained with minimal additional data to perform predictions of the time series of high temperature measurements obtained from the Engineering Test Unit (ETU) at Kairos Power. The performance of the LSTM and Transformer networks is investigated by varying the size of the lookback window and forecast horizon. The results of this study show that LSTM networks have lower prediction errors than Transformers, but LSTM errors increase more rapidly with increasing lookback window size and forecast horizon compared to the Transformer errors.

LSTM↗

Stripping off the implementation complexity of physics-based model predictive control for buildings via deep learning

Over the past decade, model predictive control (MPC) has been considered as the most promising solution for intelligent building operation. Despite extensive effort, transfer of this technology into practice is hampered by the need to obtain an accurate controller model with minimum effort, the need of expert knowledge to set it up, and the need of increased computational power and dedicated software to run it. A promising direction that tackles the last two problems was proposed by approximate explicit MPC where the optimal control policies are learned from MPC data via a suitable function approximator, e.g., a deep learning (DL) model. The main advantage of the proposed approach stems from simple evaluation at execution time leading to low computational footprints and easy deployment on embedded HW platforms. We present the energy savings potential of physics-based (also called 'white-box') MPC applied to an office building in Belgium. Moreover, we demonstrate how deep learning approximators can be used to cut the implementation and maintenance costs of MPC deployment without compromising performance. We also critically assess the presented approach by pointing out the major challenges and remaining open-research questions.

Drgona, Jan↗

GPS Measurements of Precipitable Water Vapor Can Improve Survey Calibration: A Demonstration from KPNO and the Mayall z -band Legacy Survey

Dual-band Global Positioning Satellite (GPS) measurements of precipitable water vapor (PWV) at the Kitt Peak National Observatory predict the overall per-image sensitivity of the Mayall z-band Legacy Survey (MzLS). The per-image variation in the brightness of individual stars is strongly correlated with the measured PWV and the color of the star. Synthetic stellar spectra through TAPAS transmission models successfully predict the observed PWV-induced photometric variation. We find that PWV absorption can be well approximated by a linear relationship with (airmass × PWV) 0.6 and present an update on the traditional treatment in the literature. The MzLS zero-point sensitivity in electrons s –1 varies with a normalized-mean absolute deviation of 61 mmag. PWV variation accounts 23 mmag of this zero-point variation. The MzLS per-image absolute sensitivity decreases by 40 mmag per effective mm of PWV. The overall gray offset portion of this variation is corrected by the calibration to a reference catalog. But the relative calibration error between blue (r – z < 0.5 mag) versus red (1.2 mag < r – z) stars increases by 0.3–2 mmag per effective mm of PWV. We argue that GPS systems provide more precise PWV measurements than using differential measurements of stars of different colors and recommend that observatories install dual-band GPS as a low-maintenance, low-cost, auxiliary calibration system. We extend our results of the need for well-calibrated PWV measurements by presenting the calculations of the PWV photometric impact on three science cases of interest: stellar photometry, supernova cosmology, and quasar identification and variability.

79 ASTRONOMY AND ASTROPHYSICS↗

Earth Independent Medical Operations (EIMO) Datascope: Challenges and Potential Solutions

Data flows and storage/retrieval capacity are severely constrained during missions in space and challenges will become even greater during exploration class missions. There is a need for an artificial intelligence (AI)-based clinical decision support system (CDSS) to monitor and analyze data to provide real-time consultative support for crew medical officer (CMO) decision-making. EIMO is defined as the gradual transition of medical care and decision making from terrestrial to space-based assets, enabling support of astronaut health and performance and reducing overall mission risk. While a hallmark of this paradigm shift from low-earth orbit is that on-board care will increasingly become the responsibility of the astronauts for primary management and decision making, terrestrial assets will continue to be paramount in pre-mission screening and planning, as well as prevention, health maintenance and long-term care contingencies. New capabilities and systems that enable progressively more robust and resilient systems and crews will be necessary to reduce risk and increase probability of deep space exploration mission success. An aspiration for EIMO is to develop AI-enhanced solutions for analysis of crew health & performance data and to facilitate clinical decision support for autonomous medical operations. A “system of systems” approach is envisioned whereby EIMO will deploy AI-supported natural language processing and machine learning (ML) techniques to utilize embedded reference databases and real-time data streams [input vectors] from multiple data sources. Constituent input vectors may include environmental controls, countermeasures data, behavioral data, physiologic wearables, point-of-care laboratory tests, personalized medical records, inventory trade space risk assessments, COTS medical databases, and ground support inputs. An ideal AI capability would possess trained fusion algorithms to cross reference input vectors with medical ‘knowledge’ [cultivated database] to stratify relevant data streams for predictive and actionable capabilities. In addition, EIMO will feature mobility, in that it can be accessed and can push/pull data within and between multiple vehicles/habitats. Large amounts and variable sources of data can be leveraged to diagnose, inform treatment strategies, and potentially predict medical events and performance decrements. Inclusion of advanced training tools using extended reality will enable increasingly autonomous medical care to aid a CMO when ground support is unavailable or time-delayed beyond required action window, e.g., emergent medical situations. EIMO CDSS would require very large datasets to train pre-flight and significant amounts of data are needed to support ML via in-flight CDSS operations. An additional challenge will be to find sufficient data to train a model relevant to astronaut demographics. The rapid, accelerating evolution of this field creates a propitious solution space to leverage multi-modal AI through public-private partnership(s). The status of multi-modal AI systems today would preclude their use for long duration missions as they remain unreliable and are subject to “digital hallucinations” and other errors that could pose operational risk. A federated labs structure is being considered to test and optimize data flow from the multiple input vectors leading to field testing in suitable ground/flight analogs. Critical to the success of an EIMO CDSS will be integration and interoperability and success will be defined by a system that can serve as an in-flight medical consult for the CMO providing critical support during medical contingencies. Benefits to terrestrial medicine may be significant as an outflow of the EIMO medical system, particularly for remote areas and communities lacking significant infrastructure, personnel and resources.

J Lemery↗