Search NASA⌕ Search

SEARCH · Search NASA

Results for “adaptive model predictive control”

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 397 records · Page 22

Extending Parsimonious Bayesian Inference

Parsimonious Bayesian inference is a theoretical framework for efficient data assimilation that seeks to balance increased consistency between predictions and training data against corresponding increases in model complexity. Within this framework, over-training is understood as optimization that encodes excessive information within model parameters while only achieving small improvements between predictions and training data. This project aims to develop practical methods of limiting excess model information during optimization. One key observation is that practical heuristics for parsimonious learning in high-dimensions must balance expressivity, i.e. the ability of the model to capture diverse predictions with only a few non-zero parameters, against discoverability, i.e. the ability to train the model with gradient-based optimization and drive parameters to low information states. As such, we developed logical activation functions that are able to adaptively approximate arbitrary truth tables that define Boolean logic operations within a probabilistic framework. These functions have demonstrated the ability to learn exclusive disjunction (XOR) and conditioned disjunction (if [condition] then [result_if_true] else [result_if_false]) within a single layer of a neural network. To efficiently exploit these activation functions to drive parsimonious learning required several other advances within the domain of variational inference. The most efficient form of complexity suppression is structured sparsification, driving most model parameters to zero while achieving the structural coherence among nonzeros needed for bandwidth reduction. Such models are not only far more efficient at suppressing information-theoretic complexity, they also reduce the other forms of complexity (computations, communication, storage, and the number of dependencies needed to evaluate predictions). Aiming to support enhanced sparsification, this project examined new approaches to high-dimensional variational inference that allow us to calibrate and control parameter uncertainty during optimization. By identifying which parameters can sustain sparsifying perturbations with little impact on prediction quality, we can develop better pruning strategies by framing them as approximate Bayesian inference. These advances also open paths to mitigate concerns with deploying advanced learning methods in resource-constrained environments, such as running models on power-limited or communication-limited devices.

97 MATHEMATICS AND COMPUTING↗

A Multiple Flux-tube Solar Wind Model

We present a new model, MULTI-VP, which computes the three-dimensional structure of the solar wind and includes the chromosphere, the transition region, and the corona and low heliosphere. MULTI-VP calculates a large ensemble of wind profiles flowing along open magnetic field lines that sample the entire three-dimensional atmosphere or, alternatively, a given region of interest. The radial domain starts from the photosphere and typically extends to about 30 R{sub ⊙}. The elementary uni-dimensional wind solutions are based on a mature numerical scheme that was adapted in order to accept any flux-tube geometry. We discuss here the first results obtained with this model. We use Potential Field Source-surface extrapolations of magnetograms from the Wilcox Solar Observatory to determine the structure of the background magnetic field. Our results support the hypothesis that the geometry of the magnetic flux-tubes in the lower corona controls the distribution of slow and fast wind flows. The inverse correlation between density and speed far away from the Sun is a global effect resulting from small readjustments of the flux-tube cross-sections in the high corona (necessary to achieve global pressure balance and a uniform open flux distribution). In comparison to current global MHD models, MULTI-VP performs much faster and does not suffer from spurious cross-field diffusion effects. We show that MULTI-VP has the capability to predict correctly the dynamical and thermal properties of the background solar wind (wind speed, density, temperature, magnetic field amplitude, and other derived quantities) and to approach real-time operation requirements.

79 ASTRONOMY AND ASTROPHYSICS↗

Engine identification for adaptive control

An attempt to obtain a dynamic model for a turbofan gas turbine engine for the purpose of adaptive control is described. The requirements for adaptive control indicate that a dynamic model should be identified from data sampled during engine operation. The dynamic model identified was of the form of linear differential equations with time varying coefficients. A turbine engine is, however, a highly nonlinear system, so the identified model would be valid only over a small area near the operating point, thus requiring frequent updating of the coefficients in the model. Therefore it is necessary that the identifier use only recent information to perform its function. The identifier selected minimized the square of the equation errors. Known linear systems were used to test the characteristics of the identifier. It was found that the performance was dependent on the number of data points used in the computations and upon the time interval over which the data points were obtained. Preliminary results using an engine deck for the quiet, clean, shorthaul experimental engine indicate that the identified model predicts the engine motion well when there is sufficient dynamic information, that is when the engine is in transient operation.

Leonard, R. G.↗

Finite Element Simulation of Smart Structures

Finite element equations representing the behavior of piezoelectric materials when bounded to a typical structure and used as sensors and actuators were developed. Emphasis was placed on generating sensor output equations of piezoelectric sensors and responses of a typical structure bonded with piezoelectric sensors and actuators on the basis of finite element formulation. The model can predict not only structural responses due to both mechanical and electrical loading but also electrical potential due to mechanical or thermal effects. The resulted finite element equations were then used for simple control design and performance evaluation. In the control algorithm, voltages coming out from piezoelectric sensors, which are proportional to strains at sensing locations, are taken as input. The voltages applied to the piezoelectric actuators are used as output. The feasibility of integrating control algorithm with the element routine developed herein and FEAP was demonstrated. In particular, optimal independent modal space control was implemented in a software package on the basis of finite element formulation. A rudimentary finite element-control algorithm package was also developed to evaluate the performance of candidate control laws. A few numerical simulations using the software package developed herein were given. The integrated software package will provide a design tool to address issues such as how adaptive smart systems will scale to a full size aircraft, the amount of piezoelectric materials and the powers needed to actuate it for desired performance. It will also provide a viable new structural control design concept for practical applications in large flexible structures such as aerospace vehicles and aircraft.

Cui, Y. Lawrence↗

Intraspecific Reaction Norm Variation Controls the Eco-Evolutionary Consequences of Environmental Change

As environmental change accelerates globally, understanding concurrent organismal, species, and community responses is increasingly vital. Here, we examine these collective responses by incorporating genotype-specific thermal reaction norms into an eco-evolutionary predator-prey model, allowing us to track simultaneous phenotypic, ecological, and evolutionary responses to environmental change within ecological communities. We show that the reaction norms expressed by genotypes within a population determine how a community switches between different eco-evolutionary outcomes with changes in temperature. We identify how different components of phenotypic variation in thermal reaction norms—environmental (E), additive environmental and genetic (E + G), and gene-by-environment interactions (G × E)—influence eco-evolutionary dynamics and outcomes as temperature changes. Furthermore, our findings underscore how complex eco-evolutionary responses to environmental change ultimately emerge from variation in reaction norms among genotypes, offering new mechanistic insights into environmental impacts on adaptation, the maintenance of phenotypic and genetic variation, and ecological stability, which is crucial for understanding and predicting eco-evolutionary effects of rapid environmental change in the future.

Eco-evolutionary↗

Cementation of Nuclear Wastes: From Past to Future in a Sustainable Approach - 18266

Cementation of nuclear wastes is a widespread technology due to several factors such as excellent feedback from past experience, simplicity of implementation and versatility. Indeed, as nuclear wastes come in many physical states (bulk wastes, powders, sludges), cementation is the immobilisation process that can easily accommodate this large range of wastes provided the appropriate mixing technology is used: heterogenous cementation (grouting of bulky wastes) or homogenous cementation (intermixing process between the powder or the sludge and the cementitious material). However, nuclear wastes equally present an important chemical variability. This is especially true as New Areva's range of nuclear activities covers the entire fuel cycle: mining, conversion, enrichment and recycling and dismantling and decommissioning. The chemical interactions between the waste and the cement matrix are carefully studied by New Areva's R and D teams in order to provide comprehensive solutions by developing cement formulations which are adapted to wastes characteristics. The aim of the studies is to obtain cement waste-forms that meet the waste acceptance criteria, the final product being thus in perfect adequacy with the disposal requirements. For over 30 years New Areva has been developing cement formulations adapted to the physical and chemical specificities of the wastes accounting for constrains imposed by waste characteristics. The overall approach includes the definition of the appropriate cement composition at laboratory scale, engineering design, equipment manufacturing, testing and commissioning of the cementation facilities. With this strong technical background, a specific methodology has been developed by New Areva teams in order to allow a rapid screening of possible solutions for a given waste (in terms of cement chemistry adapted to the waste) in a learn-fast approach that is based on theoretical analysis and small-scale tests. The methodology was applied for wastes presenting a complex nature (such as reactive metals, ion exchange resins, oils and solvents, mercury, etc), known to interact with Portland cement from a chemical point of view and leading to durability issues of the wasteform (swelling, strength reduction, leaching) and thus reducing the final waste loading. Once the screening process is over, the main drivers for choosing the most appropriate solution to be further developed at pilot scale are: reducing the overall volume and the impact of the final waste package, ensure the waste package performance as defined by the disposal requirements and control the long-term behaviour in final disposal. Moreover, a modelling approach is addressed, aiming to develop a thermo-chemo-mechanical (THM) concrete models that are able to predict the behaviour of the structure (waste form) in time. The aim of the present paper is to describe the main principles within the developed methodology and illustrate it by using a few specific examples. Several distinct cement formulations are developed for complex wastes, limiting the chemical interaction between the cement matrix and the given wastes, obtaining better waste loadings and better overall durability behaviour which facilitates long term behaviour demonstration. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗

A Multi-Scale Computational Platform for Predictive Modeling of Corrosion in Al-Steel Joints (Final Report)

The research team proposed to develop innovative multi-scale models to predict corrosion and the resulting mechanical performances in aluminum-steel joints. The methods of joining considered are resistance spot welding, self-piercing riveting, and rivet-welding, all suitable for mass production applications. The multi-scale models integrate high throughput first-principle calculations based on density functional theory (DFT), high throughput calculation of phase diagrams (CALPHAD) modeling, and finite element method (FEM) simulations. These models are to be validated through laboratory experiments. Furthermore, the models are available as open source so as to enable scientists and engineers in the community to adapt and contribute to the development and application. The approaches rely on the research team’s extensive experience on the prediction of properties of individual phases at finite temperatures and variable compositions through DFT calculations, and our broad expertise on dissimilar material joining and their corrosion. The proposed computational framework enables high throughput computations for improved predictions of corrosion and the associated mechanical performance in dissimilar material joints, resulting in significant reduction in computational time needed by the current state-of-the-art methods. With the participation of researchers from three universities, an auto manufacturer, two manufacturing technology/equipment suppliers, and a software developer/vendor, the interdisciplinary research team applies the technical development on both phase-based modeling and laboratory experiments into the automobile body joining processes for validation and technology demonstration. The global cost of corrosion was estimated at about 3.4% of the global GDP in 2013. By using available corrosion control practices, it is estimated a saving between 15-35% of the cost of corrosion. In the U.S., more than $276 billion is spent repairing corrosion damage. Prediction of the corrosion and its impact on performance of the dissimilar material joints is critical for reducing the massive number of the current corrosion-based recalls for automobiles. Thus, the project goal is to develop models to enable predictive maintenance and end-of-life planning of multi-metal joints with risk of corrosion under different conditions such as exposure to high temperatures in summer and salt solutions in winter, quantified through its pH. An academia-industry consortium led by the University of Michigan and including Pennsylvania State University, University of Illinois Urbana-Champaign, University of Georgia, General Motors Company, Livermore Software Technology Corporation, and Optimal Process Technologies, LLC. created multi-scale models for prediction of corrosion in aluminum-steel joint structures such of them used in vehicle subassemblies – chassis and transmission systems. Starting from the first principle calculations, the team developed mathematical and data-driven models to predict the metallic components, which are formed during joining of two metals, for example aluminum and steel - a lightweight multilateral system which is currently used in more than 60% car bodies. These models were used for simulating chemical reactions that are happening when the joining metallic components are exposed to high temperatures and different pH values. The team was able to predict how the corrosion installs on the metallic components and how they lead to a sudden failure of components in cars. Newly developed machine learning algorithms combining Science, Technology, Engineering and Math disciplines, advanced finite element simulation and experimental validations have been integrated in a platform for prediction of the corrosion evolution and prediction the failure of joints under mechanical loadings and fatigue. Moreover, based on machine learning and inverse analysis, the team proposed solutions for designing new metallic alloys less susceptible to corrosion when joining multi-material assembles. An average of 4% error compared with experiments was achieved for the most common joints that are used in vehicle subassemblies.

36 MATERIALS SCIENCE↗

SISGR: Chemomechanics of Far-From-Equilibrium Interfaces (COFFEI)

Portable, reliable, and deployable devices for energy storage and conversion require fundamental changes in design of solid-state composites comprising ceramics and metals. These materials comprise the electrodes and electrolytes of next-generation solid-oxide fuel cells and all solid-state batteries, forming solid-state functional composites. The advent of solid-state batteries – which replace liquid electrolytes with solid electrolytes capable of lithium ion transport for reliable energy storage in portable batteries – and the increased demand space for all solid-state fuel cells capable of oxygen reduction at intermediate temperatures remain important challenges for improved material stability and decreased system cost. However, little is understood about three fundamental facets of materials that enable such solid-state energy applications. First, how do such materials deform, fracture, or delaminate under operando conditions? Second, how does such mechanical deformation limit or facilitate electronic and ionic transport within and across such material interfaces? Third, how we can predictably design interface-rich composites to engineer both structural and electrochemical stability? This COFFEI Group comprised expertise from Materials Science & Engineering and Nuclear Science & Engineering to integrate unique in situ experiments, simulations, and fabricated interfaces that address these fundamental questions in solid-state interfaces of nanoscale composites that will guide solid-state electrochemistry, transport kinetics, and mechanical deformation for nonstoichiometric materials that enable such applications. In particular, we built on COFFEI’s understanding of chemomechanical coupling among defect concentrations, ionic transport, electron transport, and stored elastic energy that is particularly acute in the far-from-equilibrium conditions typical of energy device applications. By tailoring our focus to solid-state interfaces, we addressed these important issues by (a) developing and applying advanced in situ and ex situ characterization tools to characterize model materials and interfaces synthesized with molecular-level control, under both laboratory-controlled and extreme environments representative of energy device operation; and (b) employing computational modeling and simulation frameworks to predict transport mechanisms, reactivity and stability of these model materials and interfaces under significant chemical strains typical of energy device operation. Recent progress provided insights to additional materials systems and electrochemomechanical fatigue and fracture that were not fully envisioned when the program was initiated. Specifically, in the final three years of COFFEI we pursued two integrated thrusts, with complementary focus. Thrust I focused on failure-resistant electrochemomechanical composites, while Thrust II focused on strain-modulated conductivity and reactivity across interfaces. In contrast to our initial COFFEI focus, these thrusts concentrated wholly on solid-state material interfacial interactions and included greater integration of multiscale visualization including in situ electron microscopy of strained structures/interactions and mesoscale simulations. Successful development of functionally superior and long-lived battery and fuel cell systems and stress adaptable oxides requires a deeper, fundamental understanding of the coupling among the historically important subfields of solid-state electrochemistry, transport kinetics, and mechanical deformation for nonstoichiometric metal oxide electrodes. In this program, the understanding and the application of chemomechanical coupling of defect concentrations, ionic transport, electro-catalytic activity and stored elastic energy, particularly acute in the far-from-equilibrium conditions typical of energy device applications, are being refined and implications for device operation clarified, including for miniaturized solid-state batteries and fuel cells.

36 MATERIALS SCIENCE↗

Integrated Modeling Tools for Thermal Analysis and Applications

Integrated modeling of spacecraft systems is a rapidly evolving area in which multidisciplinary models are developed to design and analyze spacecraft configurations. These models are especially important in the early design stages where rapid trades between subsystems can substantially impact design decisions. Integrated modeling is one of the cornerstones of two of NASA's planned missions in the Origins Program -- the Next Generation Space Telescope (NGST) and the Space Interferometry Mission (SIM). Common modeling tools for control design and opto-mechanical analysis have recently emerged and are becoming increasingly widely used. A discipline that has been somewhat less integrated, but is nevertheless of critical concern for high precision optical instruments, is thermal analysis and design. A major factor contributing to this mild estrangement is that the modeling philosophies and objectives for structural and thermal systems typically do not coincide. Consequently the tools that are used in these discplines suffer a degree of incompatibility, each having developed along their own evolutionary path. Although standard thermal tools have worked relatively well in the past. integration with other disciplines requires revisiting modeling assumptions and solution methods. Over the past several years we have been developing a MATLAB based integrated modeling tool called IMOS (Integrated Modeling of Optical Systems) which integrates many aspects of structural, optical, control and dynamical analysis disciplines. Recent efforts have included developing a thermal modeling and analysis capability, which is the subject of this article. Currently, the IMOS thermal suite contains steady state and transient heat equation solvers, and the ability to set up the linear conduction network from an IMOS finite element model. The IMOS code generates linear conduction elements associated with plates and beams/rods of the thermal network directly from the finite element structural model. Conductances for temperature varying materials are accommodated. This capability both streamlines the process of developing the thermal model from the finite element model, and also makes the structural and thermal models compatible in the sense that each structural node is associated with a thermal node. This is particularly useful when the purpose of the analysis is to predict structural deformations due to thermal loads. The steady state solver uses a restricted step size Newton method, and the transient solver is an adaptive step size implicit method applicable to general differential algebraic systems. Temperature dependent conductances and capacitances are accommodated by the solvers. In addition to discussing the modeling and solution methods. applications where the thermal modeling is "in the loop" with sensitivity analysis, optimization and optical performance drawn from our experiences with the Space Interferometry Mission (SIM), and the Next Generation Space Telescope (NGST) are presented.

Milman, Mark H.↗

Minimizing Segregation during the Controlled Directional Solidification of Dendric Alloys

Gravity-driven convection induced in the liquid by density gradients of temperature or composition disrupts uniform dendritic growth during controlled directional solidification and promotes severe macrosegregation. The solute-rich region about the dendrite tip appears to play a pivotal role in channel initiation. Allen and Hunt referred to this region as an "initial transient" or dynamic region constituting steep concentration gradients. Experimental investigation also point to the role the tip region plays in developing microstructure. Hellawell and co-workers showed that flow-through dendritic channels could be effectively disrupted, and segregation minimized, during the gradient freezing of bulk castings by rotating the melt through a slight angle with respect to Earth's gravity vector. Adapting this principle to controlled directional solidification, it has been shown" that segregation in dendritic alloys can be minimized, and properties improved, by processing the sample near horizontal in conjunction with a slow axial rotation of the crucible. It is postulated that the observed microstructural uniformity arises by maintaining the developing solute field about the dendrite tip. Solute rejected during vertical directional solidification will rise or sink parallel to the primary dendrite arms during axial rotation setting the stage for accumulation, instabilities, and segregation. In contrast, during horizontal growth, the rejected solute will sink or rise perpendicular to the primary dendrite. Now, in the presence of a slight axial rotation, solute that was initially sinking (or rising) will find itself above (or below) its parent dendrite, i.e., still about the tip region. The following is intended to experimentally demonstrate the viability of this concept in coordination with a model that gives predictive insight regarding solute distribution about growing dendrites. Alloys based on the lead-tin eutectic system were used in this study. The system is well characterized, the constituent metals are available in a very pure form, and the thermophysical properties are well known. During solidification of hypoeutectic alloys, e.g., 55 wt pct Pb, the primary dendrites reject the less dense tin, and for the hypereutectic alloys, e.g., 75 wt pct Sn, the primary dendrites reject denser lead. Alloys were prepared by melting appropriate amounts of lead and tin in a glass crucible after which the homogeneous liquid was sucked directly into 5-mm i.d. glass tubes. The sample tube, containing approximately 30 cm of alloy, was then mechanically driven into the directional solidification furnace assembly and positioned such that approx. 20 cm of the sample was remelted. Subsequently, directional solidification was initiated by withdrawing the sample through a water-cooled jacket at a constant growth velocity of 2 ,microns/s. After 5 to 6 cm of growth, the sample was quickly removed from the furnace and quenched in a water bath to preserve the solid-liquid interface. Samples were directionally solidified vertically upward, nearly horizontally, and some in conjunction with an applied axial rotation of the crucible. Temperature gradients at the solid-liquid interface were measured with an in-siru K-type thermocouple. Solidified samples were cut perpendicular and parallel to the growth direction and conventionally prepared for microscopic examination.

Grugel, Richard N.↗

Adaptive Deep Reinforcement Learning Algorithm for Distribution System Cyber Attack Defense With High Penetration of DERs

With grid modernization, smart inverters are increasingly used to execute advanced controls for distribution network reliability. However, this also increases the cyber-attack space. Here this paper focuses on the defense approaches to restore the system to normal operation circumstances in the presence of cyber-attacks. A unique deep reinforcement learning (DRL) method is developed to minimize voltage violations and reduce power losses for impacted feeders. The defense problem is reformulated as a Markov decision-making process to dynamically control DERs while minimizing load shedding. This is achieved via an improved soft actor-critic (SAC)-based DRL algorithm, which can govern DER set points and load-shedding scenarios in discrete and continuous modes via the auto-tune entropy and Gaussian policy features. Numerical comparison results on the modified IEEE 123-node system with other control approaches, such as Volt-VAR (VV), Volt-Watt (VW), and model predictive control (MPC) show that the proposed method can eliminate voltage violations and provide feasible control actions that perform complete mitigation of cyber-threats.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Helicopter Human Factors

Even under optimal conditions, helicopter flight is a most demanding form of human-machine interaction, imposing continuous manual, visual, communications, and mental demands on pilots. It is made even more challenging by small margins for error created by the close proximity of terrain in NOE flight and missions flown at night and in low visibility. Although technology advances have satisfied some current and proposed requirements, hardware solutions alone are not sufficient to ensure acceptable system performance and pilot workload. However, human factors data needed to improve the design and use of helicopters lag behind advances in sensor, display, and control technology. Thus, it is difficult for designers to consider human capabilities and limitations when making design decisions. This results in costly accidents, design mistakes, unrealistic mission requirements, excessive training costs, and challenge human adaptability. NASA, in collaboration with DOD, industry, and academia, has initiated a program of research to develop scientific data bases and design principles to improve the pilot/vehicle interface, optimize training time and cost, and maintain pilot workload and system performance at an acceptable level. Work performed at Ames, and by other research laboratories, will be reviewed to summarize the most critical helicopter human factors problems and the results of research that has been performed to: (1) Quantify/model pilots use of visual cues for vehicle control; (2) Improve pilots' performance with helmet displays of thermal imagery and night vision goggles for situation awareness and vehicle control; (3) Model the processes by which pilots encode maps and compare them to the visual scene to develop perceptually and cognitively compatible electronic map formats; (4) Evaluate the use of spatially localized auditory displays for geographical orientation, target localization, radio frequency separation; (5) Develop and flight test control/display concepts; (6) Quantify, model, predict, and improve pilots, workload-management strategies; and (7) Design computer-game trainers to reduce training time and cost.

Hart, Sandra G.↗

BOP-301 and BOP-302R: Test Definitions and Analyses

Following the conclusion of the International Atomic Energy Agency (IAEA) coordinated research project (CRP) on analysis of two EBR-II tests, Argonne has continued its EBR-II analysis with additional tests from the Shutdown Heat Removal Test (SHRT) series. The tests selected for this follow-up analysis are BOP-301 and BOP-302R, two unprotected loss of heat of sink tests in which the intermediate sodium pump tripped without scramming the control rods or tripping the primary pumps. The IAEA CRP focused on the SHRT-17 and SHRT-45R tests, the most severe protected and unprotected loss of flow tests performed during the SHRT series. With the primary pumps tripping, flow rates decreased to levels where flowmeter uncertainty was very high. Accurately capturing the progressions of the two tests despite this uncertainty was one of the key modeling and simulation challenges during the CRP. BOP-301 and BOP-302R were selected because the primary sodium pumps did not trip for those tests. Primary flow rate measurements are known with greater certainty. Power and temperature discrepancies can be explored without concern that flow predictions match test measurements but not the actual flow rates from the tests. These two tests were performed during the same test window as SHRT-45R and had the same core load configuration. Therefore, core models created for analysis of SHRT-45R were easily adapted for analysis of BOP-301 and BOP-302R. Another reason for analyzing BOP-301 and BOP-302R is to analyze tests in which the core inlet temperature changed significantly, which did not occur during the loss of flow tests. The transient progression of the unprotected SHRT-45R test was driven by a decreasing core flow rate while the core inlet temperature remained relatively constant. For the BOP tests, the transient progression was driven by an increasing core inlet temperature while the core flow rate was stable. This report defines the BOP-301 and BOP-302R tests and provides a selection of measured test data. Initial predictions from SAS4A/SASSYS-1 simulations of these tests are then presented. Following these simulations, additional analysis was performed with the Dakota uncertainty quantification and optimization toolkit to determine which parameters have the greater impact on the transient results and the optimal values of the most important parameters.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Bayesian network ensemble as a multivariate strategy to predict radiation pneumonitis risk

Purpose: Prediction of radiation pneumonitis (RP) has been shown to be challenging due to the involvement of a variety of factors including dose–volume metrics and radiosensitivity biomarkers. Some of these factors are highly correlated and might affect prediction results when combined. Bayesian network (BN) provides a probabilistic framework to represent variable dependencies in a directed acyclic graph. The aim of this study is to integrate the BN framework and a systems’ biology approach to detect possible interactions among RP risk factors and exploit these relationships to enhance both the understanding and prediction of RP. Methods: The authors studied 54 nonsmall-cell lung cancer patients who received curative 3D-conformal radiotherapy. Nineteen RP events were observed (common toxicity criteria for adverse events grade 2 or higher). Serum concentration of the following four candidate biomarkers were measured at baseline and midtreatment: alpha-2-macroglobulin, angiotensin converting enzyme (ACE), transforming growth factor, interleukin-6. Dose-volumetric and clinical parameters were also included as covariates. Feature selection was performed using a Markov blanket approach based on the Koller–Sahami filter. The Markov chain Monte Carlo technique estimated the posterior distribution of BN graphs built from the observed data of the selected variables and causality constraints. RP probability was estimated using a limited number of high posterior graphs (ensemble) and was averaged for the final RP estimate using Bayes’ rule. A resampling method based on bootstrapping was applied to model training and validation in order to control under- and overfit pitfalls. Results: RP prediction power of the BN ensemble approach reached its optimum at a size of 200. The optimized performance of the BN model recorded an area under the receiver operating characteristic curve (AUC) of 0.83, which was significantly higher than multivariate logistic regression (0.77), mean heart dose (0.69), and a pre-to-midtreatment change in ACE (0.66). When RP prediction was made only with pretreatment information, the AUC ranged from 0.76 to 0.81 depending on the ensemble size. Bootstrap validation of graph features in the ensemble quantified confidence of association between variables in the graphs where ten interactions were statistically significant. Conclusions: The presented BN methodology provides the flexibility to model hierarchical interactions between RP covariates, which is applied to probabilistic inference on RP. The authors’ preliminary results demonstrate that such framework combined with an ensemble method can possibly improve prediction of RP under real-life clinical circumstances such as missing data or treatment plan adaptation.

60 APPLIED LIFE SCIENCES↗

Short-Term Forecasting Across a Network for the Autonomous Wind Farm: Preprint

In an autonomous wind farm, turbines will use information from nearby turbines to achieve wind farm-level objectives such as optimizing the overall performance of a wind farm, ensuring resiliency when other sensors fail, and adapting to changing local conditions. In this paper, the wind farm can be modeled as a network within which turbines (nodes) share information across designated communication channels, with a focus on turbines at the outside of the wind farm capturing local effects and sharing that information with downstream turbines. Understanding of varied inflow conditions can be especially important in complex terrain. This information can be used to monitor turbines, self-organize turbines into groups, and predict the power performance of a wind farm. In particular, this paper describes an autonomous wind farm that incorporates information from local sensors in real time to predict wind speed and wind direction at each turbine over a short-term horizon. Results indicate that the estimate of wind direction can be used to improve the knowledge of the wind speed and direction over the persistence method on a 10- 15-minute time horizon. These short-term forecasts can also be used to facilitate advanced control methods such as feedforward control within a wind farm.

17 WIND ENERGY↗

Short-Term Forecasting Across a Network for the Autonomous Wind Farm

In an autonomous wind farm, turbines will use information from nearby turbines to achieve wind farm-level objectives such as optimizing the overall performance of a wind farm, ensuring resiliency when other sensors fail, and adapting to changing local conditions. In this paper, the wind farm can be modeled as a network within which turbines (nodes) share information across designated communication channels, with a focus on turbines at the outside of the wind farm capturing local effects and sharing that information with downstream turbines. Understanding of varied inflow conditions can be especially important in complex terrain. This information can be used to monitor turbines, self-organize turbines into groups, and predict the power performance of a wind farm. In particular, this paper describes an autonomous wind farm that incorporates information from local sensors in real time to predict wind speed and wind direction at each turbine over a short-term horizon. Results indicate that the estimate of wind direction can be used to improve the knowledge of the wind speed and direction over the persistence method on a 10-15-minute time horizon. These short-term forecasts can also be used to facilitate advanced control methods such as feedforward control within a wind farm.

17 WIND ENERGY↗

ISD3: a particokinetic model for predicting the combined effects of particle sedimentation, diffusion and dissolution on cellular dosimetry for in vitro systems

The development of particokinetic models describing the delivery of insoluble or poorly soluble nanoparticles to cells in liquid cell culture systems has improved the basis for dose-response analysis, hazard ranking from high-throughput systems, and now allows for translation of exposures across in vitro and in vivo test systems. Complimentary particokinetic models that address processes controlling delivery of both particles and released ions to cells, and the influence of particle size changes from dissolution on particle delivery for cell-culture systems would help advance our understanding of the role of particles ion dosimetry on cellular toxicology. We developed ISD 3 , an extension of our previously published model for insoluble particles, by deriving a specific formulation of the Population Balance Equation for soluble particles. ISD3 describes the time, concentration and particle size dependent dissolution of particles, their delivery to cells, and the delivery and uptake of ions to cells in in vitro liquid test systems. The model is modular, and can be adapted by application of any empirical model of dissolution, alternative approaches to calculating sedimentation rates, and cellular uptake or treatment of boundary conditions. We apply the model to calculate the particle and ion dosimetry of nanosilver and silver ions in vitro after calibration of two empirical models, one for particle dissolution and one for ion uptake. The results demonstrate utility and accuracy of the ISD 3 framework for dosimetry in these systems. Total media ion concentration, particle concentration and total cell-associated silver time-courses were well described by the model, across 2 concentrations of 20 and 110 nm particles. ISD 3 was calibrated to dissolution data for 20 nm particles as a function of serum protein concentration, but successfully described the media and cell dosimetry time-course for both particles at all concentrations and time points. We also report the finding that protein content in media has effects both on the initial rate of dissolution and the resulting near-steady state ion concentration in solution.

59 BASIC BIOLOGICAL SCIENCES↗

Optimizing Hydronic Heating for Comfort and Performance in Multifamily Housing

Inefficient control settings in multifamily boilers often lead to substantial energy and cost penalties. To address this, a Fault Detection and Diagnostic (FDD) tool was developed to automate data analysis and identify operational faults such as suboptimal outdoor temperature sensor placement, misconfigured outdoor air reset (OAR) curves, excess boiler cycling, and domestic hot water (DHW) setpoint errors. By comparing pre- and post-implementation periods and applying engineering models, the tool quantifies energy savings and reduces manual analysis time by over 90%. Testing on over 100 monitored sites and a targeted subset of 12 buildings showed an average 11% energy savings from remote optimization; further validation across 19 OAR curve changes confirmed the tool’s accuracy, predicting actual savings within ±5% for most cases. Simple payback can be under three years for many multifamily buildings, though rising hardware, labor, and fuel costs create uncertainties, and decarbonization goals increasingly shift focus to electrification. The FDD tool remains invaluable for optimizing existing boilers, enhancing future electrification measures, and adapting to new technologies by refining building load estimates. In doing so, it supports both near-term efficiency and long-term transitions to low-carbon alternatives, ensuring buildings achieve substantial cost and energy benefits throughout their system lifecycles.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗