Search NASA⌕ Search

SEARCH · Search NASA

Results for “Model based development”

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 433 records · Page 24
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Assessing Geospatial and Seasonal Influences on Energy and Cost-Efficiency of Drayage Trucks

The electrification of heavy-duty vehicles is a critical pathway toward improved energy efficiency in the freight sector. The current battery electric truck technology poses several challenges to commercial vehicle operations, such as limited driving range, sensitivity to climate conditions, and long recharging times. Estimating the energy consumption of heavy-duty electric trucks is crucial to assessing the feasibility of fleet electrification and its impact on the electric grid. This article focuses on developing a model-based simulation approach to predict and analyze the energy consumption of electric trucks by considering the impact of weather and geographical conditions on vehicle road load and auxiliary components power consumption, as well as the impact these factors have on driving range. Specifically, drayage trucks employed in logistics around maritime ports are used as a case study, with consideration of seasonal climate variations and geographical characteristics at different locations. The article includes results for three major container ports within the United States, providing region-specific insights into the energy requirements and driving range of the electric drayage trucks in these regions, which will inform decision-makers in integrating electric trucks into the existing drayage operations and plan investments for electric grid development.

Sujan, Vivek [ORNL] (ORCID:0000000269882342)↗
NASA NTRS1997

Analytical Modeling of Pressure Wall Hole Size and Maximum Tip-to-Tip Crack Length for Perforating Normal and Oblique Orbital Debris Impacts

This report presents the results of a study whose objective was to develop first-principles-based models of hole size and maximum tip-to-tip crack length for a spacecraft module pressure wall that has been perforated in an orbital debris particle impact. The hole size and crack length models are developed by sequentially characterizing the phenomena comprising the orbital debris impact event, including the initial impact, the creation and motion of a debris cloud within the dual-wall system, the impact of the debris cloud on the pressure wall, the deformation of the pressure wall due to debris cloud impact loading prior to crack formation, pressure wall crack initiation, propagation, and arrest, and finally pressure wall deformation following crack initiation and growth. The model development has been accomplished through the application of elementary shock physics and thermodynamic theory, as well as the principles of mass, momentum, and energy conservation. The predictions of the model developed herein are compared against the predictions of empirically-based equations for hole diameters and maximum tip-to-tip crack length for three International Space Station wall configurations. The ISS wall systems considered are the baseline U.S. Lab Cylinder, the enhanced U.S. Lab Cylinder, and the U.S. Lab Endcone. The empirical predictor equations were derived from experimentally obtained hole diameters and crack length data. The original model predictions did not compare favorably with the experimental data, especially for cases in which pressure wall petalling did not occur. Several modifications were made to the original model to bring its predictions closer in line with the experimental results. Following the adjustment of several empirical constants, the predictions of the modified analytical model were in much closer agreement with the experimental results.

Schonberg, William P.↗
DOE OSTI2025

Learning error distribution kernel‐enhanced neural network methodology for multi‐intersection signal control optimization

Traffic congestion has substantially induced significant mobility and energy inefficiency. Many research challenges are identified in traffic signal control and management associated with artificial intelligence (AI)-based models. For example, developing AI-driven dynamic traffic system models that accurately capture high-resolution traffic attributes and formulate robust control algorithms for traffic signal optimization is difficult. Additionally, uncertainties in traffic system modeling and control processes can further complicate traffic signal system controllability. To partially address these challenges, this study presents a novel, hybrid neural network model enhanced with a probability density function kernel shaping technique to formulate traffic system dynamics better and improve comprehensive traffic network modeling and control. The numerical experimental tests were conducted, and the results demonstrate that the proposed control approach outperforms the baseline control strategies and reduces overall average delays by 11.64% on average. By leveraging the capabilities of this innovative model, this study aims to address major challenges related to traffic congestion and energy inefficiency toward more effective and adaptable AI-based traffic control systems.

Wang, Hong [Oak Ridge National Laboratory (ORNL), ↗
NASA NTRS1996

Uncertainty Modeling for Structural Control Analysis and Synthesis

The development of an accurate model of uncertainties for the control of structures that undergo a change in operational environment, based solely on modeling and experimentation in the original environment is studied. The application used throughout this work is the development of an on-orbit uncertainty model based on ground modeling and experimentation. A ground based uncertainty model consisting of mean errors and bounds on critical structural parameters is developed. The uncertainty model is created using multiple data sets to observe all relevant uncertainties in the system. The Discrete Extended Kalman Filter is used as an identification/parameter estimation method for each data set, in addition to providing a covariance matrix which aids in the development of the uncertainty model. Once ground based modal uncertainties have been developed, they are localized to specific degrees of freedom in the form of mass and stiffness uncertainties. Two techniques are presented: a matrix method which develops the mass and stiffness uncertainties in a mathematical manner; and a sensitivity method which assumes a form for the mass and stiffness uncertainties in macroelements and scaling factors. This form allows the derivation of mass and stiffness uncertainties in a more physical manner. The mass and stiffness uncertainties of the ground based system are then mapped onto the on-orbit system, and projected to create an analogous on-orbit uncertainty model in the form of mean errors and bounds on critical parameters. The Middeck Active Control Experiment is introduced as experimental verification for the localization and projection methods developed. In addition, closed loop results from on-orbit operations of the experiment verify the use of the uncertainty model for control analysis and synthesis in space.

Campbell, Mark E.↗
NASA NTRS2001

Electro-Thermal Simulation Studies of SiC Junction Diodes Containing Screw Dislocations Under High Reverse-Bias Operation

The objective of this work was to conduct a modeling study of SiC P-N junction diodes operating under high reverse biased conditions. Analytical models and numerical simulation capabilities were to be developed for self-consistent electro-thermal analysis of the diode current-voltage (I-V) characteristics. Data from GRC indicate that screw dislocations are unavoidable in large area SiC devices, and lead to changes in the SiC diode electrical response characteristics under high field conditions. For example, device instability and failures linked to internal current filamentation have been observed. The physical origin of these processes is not well understood, and quantitative projections of the electrical behavior under high field and temperature conditions are lacking. Thermal calculations for SiC devices have not been reported in the literature either. So estimates or projections of peak device temperatures and power limitations do not exist. This numerical study and simulation analysis was aimed at resolving some of the above issues. The following tasks were successfully accomplished: (1) Development of physically based models using one- and two-dimensional drift-diffusion theory for the transport behavior and I-V characteristics; (2) One- and two-dimensional heat flow to account for internal device heating. This led to calculations of the internal temperature profiles, which in turn, were used to update the electrical transport parameters for a self-consistent analysis. The temperature profiles and the peak values were thus obtainable for a given device operating condition; (3) Inclusion of traps assumed to model the presence of internal screw dislocations running along the longitudinal direction; (4) Predictions of the operating characteristics with and without heating as a function of applied bias with and without traps. Both one and two-dimensional cases were implemented; (5) Assessment of device stability based on the operating characteristics. The presence of internal non-uniformities, particularly filamentary structures, was probed and demonstrated; (6) Cause and physical origins of filamentary behavior and unstable I-V characteristics were made transparent; (7) It was demonstrated that diodes containing defects would be more prone to thermal breakdown associated with the temperature dependent decrease in the thermal conductivity; and (8) Finally, negative differential resistance (S-shaped NDR) which can potential lead to device instability and filamentary behavior was shown to occur for diodes containing a line of defects such as could be associated with a screw dislocation line.

Joshi, R. P.↗
DOE OSTI2026

Development of a deep potential model for F and CF 2 etching of Si and SiO 2

An understanding of plasma-surface interactions at increasingly smaller scales is invaluable for the development of novel technologies and processing techniques. Molecular dynamics (MD) simulations can provide insights into atomic-scale interactions, though they are restricted by the availability of interatomic potentials. Machine learning methods, such as Deep Potential Molecular Dynamics (DeepMD), provide a systematic framework for the development of accurate and flexible ab initio-based models. In this work, we develop DeepMD models for the ion-enhanced etching of Si and SiO 2 by F and CF 2 radicals. We employ an active learning process to expand the data set on which the model is trained and demonstrate its effect on the model accuracy. The DeepMD results are compared to data from classical MD simulations and experiments. Physical sputtering yields of SiO 2 by Ar + ions show good agreement with previous simulation results using conventional interatomic potentials, though the predicted depth profiles are different. Etching yields are calculated as a function of ion energy and neutral to ion flux ratio for the Ar + ion-enhanced etching of SiO 2 and Si by F atoms, as well as for etching of SiO 2 by CF 2 radicals, showing reasonable agreement with experimental data. Finally, an ion-enhanced surface kinetic model is fitted to the DeepMD etch yields, and the fitted parameters are compared to quantities computed directly from DeepMD simulations. This study illustrates how molecular dynamics simulations using machine learning potentials can provide an accurate model of etching processes relevant to device manufacturing.

Kounis-Melas, Andreas [Princeton Univ., NJ (United↗
NASA NTRS2012

Prognostics of Power Electronics, Methods and Validation Experiments

Abstract Failure of electronic devices is a concern for future electric aircrafts that will see an increase of electronics to drive and control safety-critical equipment throughout the aircraft. As a result, investigation of precursors to failure in electronics and prediction of remaining life of electronic components is of key importance. DC-DC power converters are power electronics systems employed typically as sourcing elements for avionics equipment. Current research efforts in prognostics for these power systems focuses on the identification of failure mechanisms and the development of accelerated aging methodologies and systems to accelerate the aging process of test devices, while continuously measuring key electrical and thermal parameters. Preliminary model-based prognostics algorithms have been developed making use of empirical degradation models and physics-inspired degradation model with focus on key components like electrolytic capacitors and power MOSFETs (metal-oxide-semiconductor-field-effect-transistor). This paper presents current results on the development of validation methods for prognostics algorithms of power electrolytic capacitors. Particularly, in the use of accelerated aging systems for algorithm validation. Validation of prognostics algorithms present difficulties in practice due to the lack of run-to-failure experiments in deployed systems. By using accelerated experiments, we circumvent this problem in order to define initial validation activities.

Kulkarni, Chetan S.↗
NASA NTRS1984

Dilution Jets in Accelerated Cross Flows

Results of flow visualization experiments and measurements of the temperature field produced by a single jet and a row of dilution jets issued into a reverse flow combustor are presented. The flow in such combustors is typified by transverse and longitudinal acceleration during the passage through its bending section. The flow visualization experiments were designed to examine the separate effects of longitudinal and transverse acceleration on the jet trajectory and spreading rate. A model describing a dense single jet in a lighter accelerating cross flow is developed. The model is based on integral conservation equations, including the pressure terms appropriate to accelerating flows. It uses a modified entrainment correlation obtained from previous experiments of a jet in a cross stream. The flow visualization results are compared with the model calculations in terms of trajectories and spreading rates. Each experiment is typified by a set of three parameters: momentum ratio, density ratio, and the densimetric Froude number. When injection velocities are large or densities are small, the Froude number becomes very large and hence, unimportant. Therefore, the Froude number is generally significant in describing liquid experiments but is unimportant for the gas experiments in the combustor. Agreement between test and calculated results is encouraging but tends to become poorer with increasing momentum ratio. The temperature measurements are presented primarily in the form of consecutive normalized temperature profiles. Some interpolated isothermal contours are also shown. The single jet trajectories are consistently found to be swept towards the inner wall of the bend, whether injection is from the outer or the inner wall. This behavior is explained by a drifting effect which consists of a transverse velocity component across the combustor due to the developing nature of the flow along it. Plots of lateral temperature distributions of the jet indicate that under longitudinal acceleration conditions the thermal spreading of the jet may be completely suppressed. Comparison between combustor experimental results and model calculations shows poor agreement due to the drifting effect which is not taken into consideration in the model calculations. The row of jets experiments are characterized by two additional parameters: spacing ratio and confinement parameter. The results are shown in the form of consecutive normalized temperature profiles. The confinement parameter appears to become increasingly important with decreasing spacing ratio, in particular when its effect is enhanced by the drifting phenomenon and associated pressure field. A tightly spaced row of jets injected from the inner wall, prior to the bend, is surprisingly kept attached to the inner wall in spite of the strong turning. A similar attachment for a jet injected from the outer wall is not observed.

Lipshitz, Abraham↗
NASA NTRS2009

Biophysics Representation of the Two-Hit Model of Alzheimer's Disease for the Exploration of Late CNS Risks from Space Radiation

A concern for long-term space travel outside the Earth s magnetic field is the late effects to the central nervous system (CNS) from galactic cosmic ray (GCR) or solar particle events (SPE). Human epidemiology data is severely limited for making CNS risk estimates and it is not clear such effects occur following low LET exposures. We are developing systems biology models based on biological information on specific diseases, and experimental data for proton and heavy ion radiation. A two-hit model of Alzheimer s disease (AD) has been proposed by Zhu et al.(1), which is the framework of our model. Of importance is that over 50% of the US population over the age of 75-y have mild to severe forms of AD. Therefore we recommend that risk assessment for a potential AD risk from space radiation should focus on the projection of an earlier age of onset of AD and the prevention of this possible acceleration through countermeasures. In the two-hit model, oxidative stress and aberrant cell cycle-related abnormalities leading to amyloid-beta plaques and neurofibrillary tangles are necessary and invariant steps in AD. We have formulated a stochastic cell kinetics model of the two-hit AD model. In our model a population of neuronal cells is allowed to undergo renewal through neurogenesis and is susceptible to oxidative stress or cell cycle abnormalities with age-specific accumulation of damage. Baseline rates are fitted to AD population data for specific ages, gender, and for persons with an apolipoprotein 4 allele. We then explore how low LET or heavy ions may increase either of the two-hits or neurogenesis either through persistent oxidative stress, direct mutation, or through changes to the micro-environment, and suggest possible ways to develop accurate quantitative estimates of these processes for predicting AD risks following long-term space travel.

Cucinotta, Francis A.↗
NASA NTRS2007

Hypovolemia Induced Orthostatic Hypotension in Presyncopal Astronauts and Normal Subjects Relates to Hypo-Sympathetic Responsiveness

Circulating blood volume is reduced during spaceflight, leaving astronauts hemodynamically compromised after landing. Because of this hypovolemia, crew members are able to withstand a postflight 10 minute upright tilt test only if they are able to mount a hyper-sympathetic response. Previous work from this laboratory has shown that about 30% of astronauts, primarily female, have postflight sympathetic responses to tilt that are equal to or less than their preflight responses and thus, they become presyncopal. Part of the mission of the cardiovascular lab at the Johnson Space Center is to identify susceptible crewmembers before flight so that individualized countermeasures can be prescribed. The goal of this study was to develop a ground based model of hypovolemia that could be used for this purpose We tested the hypothesis that hypovolemia alone, in the absence of spaceflight, would reproduce the landing day rate of presyncope during upright tilt in normal volunteers. Further, we hypothesized that, during hypovolemia, subjects who had sympathetic responses that were equal to or less than their normovolemic responses would become presyncopal during upright tilt tests. We studied 20 subjects, 13 male and 7 female, on two separate occasions: during normovolemia and hypovolemia. We induced hypovolemia with intravenous furosemide 40 hours prior to the experiment day, followed by a 10MEq Na diet. On the normovolemia and hypovolemia test days, plasma volume, tilt tolerance and supine and standing arterial pressure, heart rate and plasma norepinephrine levels were measured. A two factor, repeated measures analysis of variance was performed to examine the differences between group (presyncopal vs. non-presyncopal) and day (normovolemia vs. hypovolemia) effects. There were no differences in baseline arterial pressure between normovolemia and hypovolemia or between presyncopal and non-presyncopal groups, but heart rates were higher with hypovolemia in both groups (presyncopal: 70 5 bpm vs. 63 3 bpm, P = 0.003, non-presyncopal: 59 2 bpm vs. 52 2 bpm, P = 0.003). Similar to patterns reported after flight, non-presyncopal subjects had greater norepinephrine responses to tilt during hypovolemia compared to normovolemia (580 79 vs. 298 37 pg/ml, P < 0.05), but presyncopal subjects did not (180 44 vs. 145 32 pg/ml, P = NS). This new model has the potential to accelerate the development of countermeasures and save flight resources. It can be used to identify astronauts who will become presyncopal on landing day, so that prospective, individualized countermeasures can be developed. In addition, it can also be used to screen candidate countermeasures prior to requests for bed rest or inflight resources.

Meck, Janice V.↗
NASA NTRS1979

Development of a model and computer code to describe solar grade silicon production processes

Mathematical models, and computer codes based on these models were developed which allow prediction of the product distribution in chemical reactors in which gaseous silicon compounds are converted to condensed phase silicon. The reactors to be modeled are flow reactors in which silane or one of the halogenated silanes is thermally decomposed or reacted with an alkali metal, H2 or H atoms. Because the product of interest is particulate silicon, processes which must be modeled, in addition to mixing and reaction of gas-phase reactants, include the nucleation and growth of condensed Si via coagulation, condensation, and heterogeneous reaction.

Srivastava, R.↗
NASA NTRSDate not supplied

Materials Informatics at NASA GRC: Machine Learning Surrogate Modeling, Data Management, and Integrated Toolsets for Establishing/Maintaining the Digital Thread

Integrated Computational Materials Engineering (ICME) has recently received widespread attention due to its promises in reducing dependence on physical testing for engineering design by relying on simulation, reducing both time and cost to market for various applications. ICME however requires validated multiscale material models, which heavily depend on available test data with full material and test pedigree, including material processing, test and measurement equipment, raw data collection, and analysis methodology and results that is findable and usable, along with integrated, efficient toolsets for effectively passing information across various length and time scales across such models. At the NASA Glenn Research Center under the Transformational Tools and Technologies Project, significant recent efforts have been directed towards establishing the required cyberinfrastructure to enable optimized ICME processes and the design of “fit-for-purpose” materials to achieve the goals outlined in the NASA Vision 2040 report. Such efforts include development of multiscale physics-based material models, which can be used to train highly efficient surrogate machine learning models, development of best practices and infrastructure for effective, traceable materials information management, and development of toolsets that integrate with physics-based codes, machine learning models, and an information management system to enable high throughput of materials data collection and analysis, establishment of digital twins and the digital thread, and automation of the ICME design process for material optimization.

Machine Learning↗
DOE OSTI2024

Rationalizing Acidic Oxygen Evolution Reaction over IrO 2 : Essential Role of Hydronium Cation

Abstract The development of active, stable, and more affordable electrocatalysts for acidic oxygen evolution reaction (OER) is of great importance for the practical application of electrolyzers and the advancement of renewable energy conversion technologies. Currently, IrO 2 is the only catalyst with high stability and activity, but a high cost. Further optimization of the catalyst is limited by the lack of understanding of catalytic behaviors at the acid‐IrO 2 interface. Here, in strong interaction with the experiment, we develop an explicit model based on grand‐canonical density function theory (GC‐DFT) calculations to describe acidic OER over IrO 2 . Compared to the explicit models reported previously, hydronium cations (H 3 O + ) are introduced at the electrochemical interface in the current model. As a result, a variation in stable IrO 2 surface configuration under the OER operating condition from previously proposed complete *O‐coverage to a mixture coverage of *OH and *O is revealed, which is well supported by in situ Raman measurements. In addition, the accuracy of predicted overpotential is increased in comparison with the experimentally measured. More importantly, an alteration of the potential limiting step from previously identified *O→*OOH to *OH→*O is observed, which opens new opportunities to advance the IrO 2 ‐based catalysts for acidic OER.

Mou, Tianyou↗
DOE OSTI2024

Rationalizing Acidic Oxygen Evolution Reaction over IrO 2 : Essential Role of Hydronium Cation

The development of active, stable, and more affordable electrocatalysts for acidic oxygen evolution reaction (OER) is of great importance for the practical application of electrolyzers and the advancement of renewable energy conversion technologies. Currently, IrO 2 is the only catalyst with high stability and activity, but a high cost. Further optimization of the catalyst is limited by the lack of understanding of catalytic behaviors at the acid-IrO 2 interface. Here, in strong interaction with the experiment, we develop an explicit model based on grand-canonical density function theory (GC-DFT) calculations to describe acidic OER over IrO 2 . Compared to the explicit models reported previously, hydronium cations (H 3 O + ) are introduced at the electrochemical interface in the current model. As a result, a variation in stable IrO 2 surface configuration under the OER operating condition from previously proposed complete *O-coverage to a mixture coverage of *OH and *O is revealed, which is well supported by in situ Raman measurements. In addition, the accuracy of predicted overpotential is increased in comparison with the experimentally measured. More importantly, in this study, an alteration of the potential limiting step from previously identified *O→*OOH to *OH→*O is observed, which opens new opportunities to advance the IrO 2 -based catalysts for acidic OER.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Unraveling Exciton Trap Dynamics and Nonradiative Loss Pathways in Quantum Dots via Atomistic Simulations

Surface defects in colloidal quantum dots are a major source of nonradiative losses, yet the microscopic mechanisms underlying exciton trapping and recombination remain elusive. Here, we develop a model Hamiltonian based on atomistic electronic calculations to investigate exciton dynamics in CdSe/CdS core/shell QDs containing a single hole trap introduced by an unpassivated sulfur atom. By systematically varying the defect depth and reorganization energy, we uncover how defect-induced excitonic states mediate energy relaxation pathways. Our simulations reveal that a single localized defect can induce a rich spectrum of excitonic states, leading to multiple dynamical regimes, from slow, energetically off-resonant trapping to fast, cascaded relaxation through in-gap defect states. Crucially, we quantify how defect-induced polaron shifts and exciton-phonon couplings govern the balance between efficient radiative emission and rapid nonradiative decay. These insights clarify the microscopic origin of defect-assisted loss channels and suggest pathways for tailoring QD optoelectronic properties via surface and defect engineering.

Defects↗
DOE OSTI2025

Advances in geophysical forensic event monitoring

Forensic analysis of man-made, non-nuclear events (such as industrial accidents, explosion experiments and mine collapses) has become more frequent and detailed owing to advancements in geophysical monitoring. Here, in this Technical Review, we demonstrate how geophysical forensic monitoring using seismic, infrasound and hydroacoustic recordings provides insights on events in the solid earth, atmosphere and underwater. Advanced techniques, including machine-learning-based models, have been developed to detect, identify and investigate these events, providing information on location, subevents, sources and explosive yield. The increase in data availability, application of advanced methods and computation and the growth of multitechnology approaches have increased the accuracy of forensic event analysis and enabled more realistic characterization of uncertainties. For example, the 2020 Beirut explosion in Lebanon demonstrated that various seismic, acoustic and other methods could be used to estimate explosive yield (and yield uncertainties) of about 1 ktonne, providing confidence in the application of these methods to smaller events where data are available. However, forensic investigations remain largely limited to known events with identified sources. Increased access to data, sophisticated analysis methods and high-resolution earth models will improve forensic event analysis further, enabling civil and scientific applications, such as localization in the search for the lost ARA San Juan submarine.

geophysics↗
DOE OSTI2024

Impact of various DIII-D diagnostics on the accuracy of neural network surrogates for kinetic EFIT reconstructions

Abstract Kinetic equilibrium reconstructions make use of profile information such as particle density and temperature measurements in addition to magnetics data to compute a self-consistent equilibrium. They are used in a multitude of physics-based modeling. This work develops a multi-layer perceptron (MLP) neural network (NN) model as a surrogate for kinetic Equilibrium Fitting (EFITs) and trains on the 2019 DIII-D discharge campaign database of kinetic equilibrium reconstructions. We investigate the impact of including various diagnostic data and machine actuator controls as input into the NN. When giving various categories of data as input into NN models that have been trained using those same categories of data, the predictions on multiple equilibrium reconstruction solutions (poloidal magnetic flux, global scalars, pressure profile, current profile) are highly accurate. When comparing different models with different diagnostics as input, the magnetics-only model outputs accurate kinetic profiles and the inclusion of additional data does not significantly impact the accuracy. When the NN is tasked with inferring only a single target such as the EFIT pressure profile or EFIT current profile, we see a large increase in the accuracy of the prediction of the kinetic profiles as more data is included. These results indicate that certain MLP NN configurations can be reasonably robust to different burning-plasma-relevant diagnostics depending on the accuracy requirements for equilibrium reconstruction tasks.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Real-time capable modeling of ICRF heating on NSTX and WEST via machine learning approaches

Abstract A real-time capable core Ion Cyclotron Range of Frequencies (ICRF) heating model on NSTX and WEST is developed. The model is based on two nonlinear regression algorithms, the random forest ensemble of decision trees and the multilayer perceptron neural network. The algorithms are trained on TORIC ICRF spectrum solver simulations of the expected flat-top operation scenarios in NSTX and WEST assuming Maxwellian plasmas. The surrogate models are shown to successfully capture the multi-species core ICRF power absorption predicted by the original model for the high harmonic fast wave and the ion cyclotron minority heating schemes while reducing the computational time by six orders of magnitude. Although these models can be expanded, the achieved regression scoring, computational efficiency and increased model robustness suggest these strategies can be implemented into integrated modeling frameworks for real-time control applications.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗