Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “generalized 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 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Scalable Hybrid Classification-Regression Solution for High-Frequency Nonintrusive Load Monitoring

Residential buildings with the ability to monitor and control their net-load (sum of load and generation) can provide valuable flexibility to power grid operators. We present a novel multiclass nonintrusive load monitoring (NILM) approach that enables effective net-load monitoring capabilities at high-frequency with minimal additional equipment and cost. The proposed machine learning based solution provides accurate multiclass state predictions while operating at a faster timescale (able to provide a prediction for each 60- Hz ac cycle used in US power grid) without relying on event-detection techniques. We also introduce an innovative hybrid classification-regression method that allows for the prediction of not only load on/off states but also individual load operating power levels. A test bed with eight residential appliances is used for validating the NILM approach. Results show that the overall method has high accuracy, good scaling and generalization properties.

feature extraction↗
DOE OSTI2026

Effects of plasma squareness on RMP response in MAST-U

The effect of the squareness of the plasma boundary shape on the plasma response to the applied n = 2 (n is the toroidal mode number) resonant magnetic perturbation, and consequently on controlling the edge-localized mode (ELM) in MAST-U in the double-null configuration, is numerically systematically investigated. The modeling is performed utilizing three different plasma response models: the single-fluid model, the two-fluid model and the magnetohydrodynamic-kinetic hybrid model. The effect of the plasma squareness on ELM control is found to depend on the equilbrium kinetic profiles. Fixing the equilibrium radial profiles to that from a recent MAST-U discharge, the experimental plasma shape is found to be close to optimal for ELM control in terms of squareness. The optimal coil phasing does not significantly vary with varying plasma squareness. These findings robustly hold with (i) two figures of merit (FoMs) defined for predicting the ELM control feasibility, and (ii) three plasma response models employed for evaluating these FoMs.

MAST-U↗
DOE OSTI2022

Reinforcement Learning for Intelligent Building Energy Management System Control *

A building energy management system (BEMS) is a computer-based system designed to monitor and control a building's energy needs. Modern BEMS rely on the sensing and connectivity capabilities of Internet of Things (IoT) technology to intelligently adjust the energy consumption to reduce cost while respecting the consumers' preferences. Increasingly, control decisions are made based on predictions by models trained using supervised machine learning methods, which still requires control policies to be formulated in a rule-based fashion. When using reinforcement learning (RL) instead, control policies are learned by observing the utility in terms of cost and comfort associated with actions such as a change in the heating system's setpoint. The resulting RL-based controllers can capture not only the dynamics of the building and the associated electrical devices, but also fluctuations in electricity prices and user demand, avoiding the need to combine multiple predictive models with tailored control policies. This chapter will provide an overview of RL-based approaches for BEMS. After sketching the taxonomy of general RL methods, we discuss the implications of relying on the individual methods in a BEMS context. Existing work applying RL is presented along the key devices controlled by BEMS systems. Finally, we summarize the state-of-the-art and sketch limitations and open research directions.

Kotevska, Olivera↗
DOE OSTI2024

Control over Conformational Landscapes of Polypeptoids by Monomer Sequence Patterning

The ability to program chain conformation and structure through control over the monomer sequence of synthetic polymers has broad implications for next-generation material design. While related problems of protein-folding and de novo design have generated accurate predictions of 3D folded chain structures, generalization to synthetic polymers remains intractable due to the requirement of large structural databases and the intrinsically disordered nature of polymer building blocks. In this work, polypeptoids, a class of peptidomimetic synthetic polymers, are utilized to build a general workflow for the study of relationships between monomer sequence and dynamic 3D chain structure in solution. Furthermore, this work demonstrates how control over the monomer sequence can alter the conformational landscape of synthetic polymers to deviate dramatically from classical chain statistics. Specifically, the distribution of end-to-end distances, as measured by double electron-electron resonance spectroscopy in dilute solvent, is systematically skewed towards shorter distances with an increasing number of hydrophobes and further refined by hydrophobe arrangement in amphiphilic polypeptoid chains.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Innovating the next generation of commercial smart building software

Nearly 30% of commercial building energy use is wasted due to equipment faults and HVAC controls problems. The result is increased emissions, compromised comfort and productivity, and less reliable coordination of building power needs with a clean grid. The energy impact alone represents $17 billion in potential savings. Today’s smart building software provides a robust solution to address these operational deficiencies. Energy management and information systems (EMIS) are saving up to 9% on average, with two-year paybacks. They are being incorporated into energy management processes, commissioning services, and utility programs. As effective as they are, two barriers prevent even deeper benefits; limited personnel to fix problems once they are identified, and the expense and time to manually implement changes in control systems. In partnership with the research community, the EMIS industry is developing new capabilities to overcome these barriers. Moving beyond siloed products for either fault detection and diagnostics, or optimal control, these new capabilities empower users to not only automatically identify faults, but also to push corrective action, and control improvements to their buildings. In this paper, several areas for enhancements are documented: ‘one-time’ correction of faults such as setpoints, schedules, and economizer lockouts; short-term active testing for automated proportional integral derivative (PID) loop tuning and functional testing; and continuous supervisory control for demand flexibility and year-round efficiency. Results are presented from a pair of partner implementations out of a dozen providers integrating these enhancements into their products, including field tests from across the country, and insights into operator acceptance and integration into operations and maintenance practices.

Casillas, Armando↗
DOE OSTI2023

Soliton formation and topology manipulation of coupled spins via ultrafast re-magnetization

The major goal of the project was to explore the properties of magnetic order driven far out of equilibrium by optical excitations. These include fundamental questions related to the interplay of magnetic, structural and electronic degrees of freedom in materials that are optically excited. These problems break down in short term quenching of the magnetization as a results of the energy of the optical pulse studying how the energy of the optical pulse and the angular momentum of the magnetization flow between the different degrees of freedom. This is followed by the longer time re-emergence of the magnetism as the system cools. Within this goal we explored the formation of solitons, both with and without non-trivial topology, via rapid re-magnetization processes after optical-driven ultrafast demagnetization. Guided by theory, we predict that turbulence and modulational instabilities will drive the formation of solitons, including dispersive shock waves, magnon droplets, and skyrmions. In the case of topological defect formation, i.e. skyrmion generation, related processes have long been predicted based on general principles of symmetry-breaking phase transitions; the density of topological defects in a long-range-ordered phase can be controlled by varying the quench rate through the second-order phase transition, i.e. the Kibble-Zurek (KZ) mechanism. We, more broadly, had the goal to study a range of other emergent magnetic behaviors after optical excitation. While great attention has been given to ultrafast demagnetization, less is known of the subsequent spin dynamics and coupling to the lattice. What is increasingly appreciated is that ultrafast demagnetization leads to spin currents that can carry angular momentum from the rapidly demagnetized sample. These spin currents appear in many ways but fundamentally controls demagnetization, drives interactions between regions of the material, and control the spin resulting structure. They can be probed many ways including THz emission which, in turn, gives insight into the demagnetization processes. Simultaneously if there is magneto-elastic coupling ultrafast demagnetization can drive structural excitations that are not expected from thermal energy added to the lattice. We explored a range of complex phenomena that arise from photo-excitation of magnetic systems with the coupling of electronic, magnetic and structural degrees of freedom. We have made a number of fundamental discoveries that are reflected in our publications list with additional work still being prepared for publication. Highlights of this work include (i) ultra-efficient, nonlinear THz surface acoustics (ii) spin-current-mediated rapid magnon localization and coalescence, (iii) spin-wave soliton formation in ferromagnetic FePt nanoparticles, (iv) dynamic phonon coupling in elemental antiferromagnetic Cr, (v) THz emission from Co/Pt bilayers and FeRh/Pt bilayers, (vi) theoretical investigation of spin hydrodynamics, solitons and shock waves, and (vii) ultrafast perturbation of magnetic domains by optical pumping.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Energy transfer between localized emitters in photonic cavities from first principles

Radiative and nonradiative resonant couplings between defects are ubiquitous phenomena in photonic devices used in classical and quantum information technology applications. In this work, we present a first-principles approach to enable quantitative predictions of the energy transfer between defects in photonic cavities, beyond the dipole-dipole approximation and including the many-body nature of the electronic states. As an example, we discuss the energy transfer from a dipolelike emitter to an 𝐹 center in MgO in a spherical cavity. We show that the cavity can be used to controllably enhance or suppress specific spin-flip and spin-conserving transitions. Specifically, we predict that an ∼10–100 enhancement in the resonant energy transfer rate can be gained in the case of the 𝐹 center in MgO at ∼10 nm distances from a dipolar source, using rather moderate cavity with quality factor 𝑄 ∼ 400. We also show that a similar suppression in the transfer rate can be achieved by off-tuning the cavity resonance relative to the emitter transition energy. The framework presented here is general and readily applicable to a wide range of devices where localized emitters are embedded in microspheres, core-shell nanoparticles, and dielectric Mie resonators. Hence, our approach paves the way to predict how to control energy transfer in quantum memories and in ultrahigh-density optical memories, and in a variety of quantum information platforms.

First-principles calculations↗
DOE OSTI2023

Model predictive control of mixing controlled compression ignition operation for low reactivity fuels

Using gasoline or other low reactivity fuels with a pilot injection or port fuel injection in a compression ignition engine has shown great potential in reducing NOx emissions while keeping high thermal efficiency compared to diesel. However, excessive combustion noise is caused by a high maximum pressure rise rate in the cylinder due to the higher fractions of premixed charge of the low-reactivity fuel. This noise can result in structural damage to engine components and as such, combustion noise limits the range of the operating parameters and makes the control of such engines challenging. In this study, a simulation environment was built up in MATLAB/Simulink leveraging a physics-based zero-dimension combustion model to capture the in-cylinder pressure time traces as well as metrics relevant to thermal efficiency and combustion noise. Here, in order to also facilitate the control of emissions, machine learning models were investigated to capture NOx emissions. A kernel-based extreme learning machine (K-ELM) performed best and had a coefficient of correlation (R-squared) of 0.998. The combustion and NOx emission models are valid for not only conventional gasoline fuel but also oxygenated alternative fuel blends at three different pilot injection strategies. In order to track key combustion metrics while keeping noise and emissions within constraints, a model predictive control (MPC) was applied for a compression ignition engine operating with a range of potential fuels and fuel injection strategies. The MPC is validated under different scenarios, including a load step change, fuel type change, and injection strategy change, with proportional–integral (PI) control as the baseline. The simulation results show that MPC reduces about 26% of ringing intensity in the transient process and 17% at the steady state for E30. Generally, MPC can optimize the overall performance through modifying the main injection timing, pilot fuel mass, and exhaust gas recirculation (EGR) fraction.

42 ENGINEERING↗
DOE OSTI2024

A dynamic volumetric heat source model for laser additive manufacturing

Melt pool scale models of laser powder bed fusion (LPBF) offer insights into the process-structure-property relationships in additive manufacturing (AM). These models often neglect physical phenomena such as vapor cavity formation and fluid mechanics to reduce computational demands. Instead, volumetric heat source models are used to represent the effects that these phenomena have on the predicted melt pool dimensions. Generally, the dimensions and effective absorption of the volumetric heat source are calibrated to reproduce melt pool dimensions observed in metallographic cross sections taken from single-track experiments on bare plate. However, the transient nature of LPBF often deviates the melt pool dimensions from the assumed steady-state conditions of single-track experiments, motivating the need for a volumetric heat source model that more generally considers the dynamic relationship between melt pool shape and laser-material interactions. Here, we introduce a two-parameter volumetric heat source model that integrates several existing models into a generalized mathematical expression, providing independent control over the radial heat distribution via the parameter k and the volumetric shape of the heat source via the parameter m. This parameterization enables the calibration of melt pool shape predictions through simultaneous adjustment of these parameters, while keeping the radial heat source dimensions consistent with the experimental spot size (D4σ) and constraining the heat source depth and absorption to physically derived expressions for cavities. Consequently, the proposed volumetric heat source model adapts to changes in the local melt pool conditions due to scanning strategy and part geometry by dynamically adjusting the heat source depth and absorption. We demonstrate the capabilities of the proposed model through comparisons with a collection of experiments from the Additive Manufacturing Benchmark (AMBench).

36 MATERIALS SCIENCE↗
DOE OSTI2025

Prediction of defect properties in concentrated solid solutions using a Langmuir-like model

The alleged existence of sluggish diffusion in high-entropy alloys has drawn controversy. In high-entropy alloys and, in general, in all solids, transport properties are controlled by point defect concentration, which must be known before performing atomistic simulations to compute transport coefficients. In this work, we present a general Langmuir-like model for defect concentration in an arbitrarily complex solid solution and apply this model to generate expressions for concentrations of vacancies and small interstitial atoms. We then calculate the vacancy concentration as a function of temperature in the equiatomic CoNiCrFeMn and FeAl alloys with modified embedded-atom-method potentials for various chemical orderings, showing there is no clear correlation between vacancy thermodynamics and chemical ordering in the CoNiCrFeMn alloy, but clear systematic patterns for FeAl. We believe this is due to the high stability of disordered, random, and ordered intermetallic phases, respectively, in the CoNiCrFeMn and FeAl systems. Finally, this work provides future avenues to the prediction of thermal interstitials and vacancies in solid solutions, which is necessary for models of nonequilibrium behavior of solid solutions.

composition↗
DOE OSTI2023

Manipulating Retained Austenite Fraction and Stability With Controlled Chemical Heterogeneities in Q&P Steels

Advanced high-strength steels go through complex heat treatment processes with the goal of retaining austenite to enhance the mechanical properties. Martensite start temperature predictions are generally based on bulk composition when designing heat treatments, so compositional heterogeneities are not considered. Here, the effects of compositional heterogeneities on retained austenite fractions in a quenching and partitioning steel are investigated. In conclusion, the results show that Mn banding from solidification can be used to change retained austenite fractions.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Ecosystems and Networks Integrated with Genes and Molecular Assemblies (ENIGMA). Final report

The goal of the Stahl lab is to use a combination of isotopic, genomic, and activity-based analyses in the field, of selected field isolates, and defined synthetic communities to identify sources and sinks of different nitrogen species in the field, focusing on the FRC at the ORNL. The most general research objective is to characterize variants in the pathway for denitrification among field isolates and synthetic microbial assemblies as a foundation for predicting the biotic and abiotic controls of nitrogen and carbon transformations in the field. Specifically, we have focused on characterizing environmental controls of nitrous oxide emissions, an important greenhouse gas and ozone depleting molecule. Studies have mapped the flux of nitrous oxide and controlling genetic elements within the FRC subsurface, indicating the importance of different enzyme variants in nitrous oxide production and consumption in controlling surface emissions. We have developed methods to quantify and catalog the isotopic signatures imparted on nitrous oxide by different nitrous oxide reductase variants in isolated organisms and model synthetic communities. We have developed methods to quantify and catalog the nitrous oxide affinity of field relevant isolates and their biomass yield when respiring nitrous oxide on different carbon substrates. We continue to expand the collection of field-relevant microbial isolates within ENIGMA by focused enrichment and isolation from field material. These objectives have been facilitated by the development of new technologies, including: 1) the development of a multiplexed automated optical density measurement system for quantifying growth kinetics of microorganisms dependent on gaseous substrates in monoculture or gas exchange in synthetic communities and 2) the development of a microcalorimetric analysis format to quantify the impact of environmental stress on microbial maintenance energy. Since field relevant processes are mediated by assemblies of organisms interacting with each other and their local environment, we have also collaborated with other ENIGMA investigators on the assembly and characterization of synthetic communities. These studies have evaluated the role of organisms lacking a complete pathway for denitrification in the control of field processes, measuring process rates of assemblies of organisms that individually lack the full denitrification pathway but mediate the complete denitrification when growing together. These efforts have so far resulted in the publication of 18 peer reviewed papers ranging from ecological and physiological drivers of pathway optimization to field-scale process characterization.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Development and assessment of prognosis digital twin in a NAMAC system

The nearly autonomous management and control (NAMAC) system is a comprehensive control system to assist plant operations by furnishing control recommendations to operators. Prognosis digital twin (DT-P) is a critical component in NAMAC for predicting action effects and supporting NAMAC decision-making during normal and accident scenarios. To quantifying and reducing uncertainty of machine-learning-based DT-Ps in multi-step predictions, this work investigates and derives insights from the application of three techniques for optimizing the performance of DT-P by long short-term memory recurrent neural networks, including manual search, sequential model-based optimization, and physics-guided machine learning. Finally, sequential model-based optimization and physics-guide machine learning result in smallest errors when the predicting transients are similar to the training data.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2022

MODELING AND PARAMETRIC STUDY OF END-GAS AUTOIGNITION TO ALLOW THE REALIZATION OF ULTRA-LOW EMISSIONS, HIGH-EFFICIENCY HEAVY-DUTY SPARK-IGNITED NATURAL GAS ENGINES

Engine knock and misfire are barriers to pathways leading to high-efficiency Spark-Ignited (SI) Natural Gas (NG) engines. The general tendency to knock is highly dependent on engine operating conditions and the fuel reactivity. The problem is further complicated by the low emission limits and the wide range of chemical reactivity in pipeline-quality natural gas. Depending on the region and the source of the natural gas, its reactivity, described by its Methane Number (MN), which is analogous to the Octane Number for liquid SI fuels, can span from 65 to 95. In order to realize diesel-like efficiencies, SI NG engines must be designed to operate at high Brake Mean Effective Pressures (BMEP), near or beyond knock limits, over a wide range of fuel reactivity. This requires a deep understanding of the combustion-engine interactions pertaining to flame propagation and End-Gas Autoignition (EGAI), i.e., the autoignition of the unburned gas (end gas) ahead of the flame front. However, EGAI, if controlled, provides an opportunity to increase SI NG engine efficiency by increasing the combustion rate and the total fraction of burned fuel, mitigating the effects of the slow flame speeds characteristic of natural gas fuels, which generally reduce BMEP and increase unburned hydrocarbon emissions. For this reason, to realize diesel-like efficiencies and ultra-low emissions on SI NG engines, this work proposes the study of the main parameters influencing the modeling and prediction of NG EGAI to allow for its control. In this work, a novel EGAI detection and onset determination method was developed to reliably quantify EGAI for data analysis and engine control. The new method allowed the prediction of EGAI on SI NG engines without the need to use engine- and operating-condition-dependent thresholds and reduced the error in quantifying the fraction of the total energy released by the EGAI event by up to 40%pts. One- and three-dimensional engine models were then developed to study the engine/fuel interactions that lead to NG EGAI and its performance benefits. These models, although having decent agreement with experimental data, showed the need to account for NOx chemistry when predicting NG EGAI due to a consistently later prediction of the EGAI onset (~1.65 crank-angle degrees) and thus, a new reduced chemical mechanism for real NG fuels was developed containing NOx chemistry. The new reduced mechanism improved the EGAI onset prediction agreement to within ±0.5 crank-angle degrees and decreased simulation time during combustion by nearly 50% when using the further reduced AREIS50NOx chemical mechanism. These models were then used to study the role of NG composition on EGAI, evaluate the engine/fuel interactions leading to NG EGAI, and perform engine optimization while leveraging EGAI to increase thermal efficiency. Piston design optimization combined with a Controlled EGAI (C-EGAI) combustion mode allowed a Heavy-Duty (HD) SI NG engine to operate at diesel-like efficiencies, i.e., Brake Thermal Efficiency (BTE) ≥44%. Experimental and modeling data analysis revealed that earlier and faster heat release increases combustion efficiency by an average of 1%pts, increases work transferred to the piston resulting in a decrease in exhaust losses by 50% depending on the engine operating condition while slightly increasing heat losses. Finally, the simulation results revealed an opportunity to further enhance the BTE (up to 50%) by enabling C-EGAI combustion at leaner conditions, λ=1.4-1.6.

Bestel, Diego Bernardi↗
DOE OSTI2023

Test of Causal Nonlinear Quantum Mechanics by Ramsey Interferometry with a Trapped Ion

Quantum mechanics requires the time evolution of the wave function to be linear. While this feature has been associated with the preservation of causality, a consistent causal nonlinear theory was recently developed. Interestingly, this theory is unavoidably sensitive to the full physical spread of the wave function, rendering existing experimental tests for nonlinearities inapplicable. Here, in this work, using well-controlled motional superpositions of a trapped ion, we set a stringent limit of 5.4×10 –12 on the magnitude of the unitless scaling factor $\tilde{ε}_γ$ for the predicted causal nonlinear perturbation.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2025

Assessing time-dependent temperature profile predictions using reduced transport models for high performing NSTX plasmas

Time-dependent, predictive simulations were performed with the 1.5D tokamak integrated modeling code TRANSP on a large set of well-analyzed, high performing discharges from the National Spherical Torus Experiment (NSTX) in order to evaluate how well modern reduced transport models can reproduce experimentally observed temperature profiles in spherical tokamaks. Overall, it is found that simulations using the Multi-Mode Model (MMM) more consistently agree with the NSTX observations than those using the Trapped Gyro-Landau Fluid (TGLF) model, despite TGLF requiring orders of magnitude greater computational cost. When considering all examined discharges, MMM has median overpredictions of electron temperature (T e ) and ion temperature (T i ) profiles of 28% and 27%, respectively, relative to the experiment. TGLF overpredicts T e by 46%, with much larger variance than MMM, and underpredicts T i by 25%. As the ratio of kinetic to magnetic field pressure (β) is increased across NSTX discharges, TGLF predicts lower T e and significant flattening of the T i profile, conflicting with NSTX observations. When using an electrostatic version of TGLF, both T e and T i are substantially overpredicted, underscoring the importance of electromagnetic turbulence in the high β spherical tokamak regime. Additionally, calculations with neural net surrogate models for TGLF were performed outside of TRANSP with a time slice flux matching transport solver, finding better agreement with experiment than the TRANSP simulations, highlighting the impact of different transport solvers and simulation techniques. Furthermore, the reasonable agreement with experiment of temperature profiles predicted by MMM motivates a more detailed examination of the sensitivities of the TRANSP simulations with MMM to different NSTX plasma regimes in a companion paper (Lestz et al Plasma Phys. Control. Fusion 67 105030), in preparation for self-consistent, time-dependent predictive modeling of NSTX-U scenarios.

Lestz, J. B. [General Atomics, San Diego, CA (Unit↗
DOE OSTI2025

Machine learning modeling and model predictive control of a closed-circuit reverse osmosis system

Closed-circuit reverse osmosis (CCRO) offers a flexible and energy-efficient alternative to conventional reverse osmosis by operating in a semi-batch mode that recycles brine, enabling higher recovery rates and reduced specific energy consumption (SEC). However, developing accurate, system-level dynamic models for CCRO remains challenging due to its nonlinear, multi-phase operation and sensitivity to variable feed water conditions. Traditional modeling approaches, such as NARMAX (nonlinear autoregressive moving average with exogenous inputs), often struggle to generalize across varying inlet feed concentrations, necessitating frequent parameter re-estimation and limiting their utility for real-time control applications. To address these limitations, we developed a long short-term memory (LSTM) neural network model trained on an extensive experimental data set from a CCRO pilot plant. The model accepts three inputs, feed flow rate, recirculation flow rate, and initial feed conductivity, and predicts three key outputs: reject conductivity, feed pump power draw, and recirculation pump power draw. We validated the LSTM model against experimental data, demonstrating its ability to distinguish between different feed conductivities and adapt to variable flow rates. Subsequently, we incorporated the LSTM model within a nonlinear model predictive control (MPC) scheme and conducted closed-loop simulations to optimize the integrated SEC (iSEC). In conclusion, the results project up to a 6% reduction in iSEC by using MPC to optimize performance over the entire experiment duration, without requiring any random excitation for data collection or parameter re-estimation.

Desalination↗
DOE OSTI2024

EFIT-Prime: Probabilistic and physics-constrained reduced-order neural network model for equilibrium reconstruction in DIII-D

We introduce EFIT-Prime, a novel machine learning surrogate model for EFIT (Equilibrium FIT) that integrates probabilistic and physics-informed methodologies to overcome typical limitations associated with deterministic and ad hoc neural network architectures. EFIT-Prime utilizes a neural architecture search-based deep ensemble for robust uncertainty quantification, providing scalable and efficient neural architectures that comprehensively quantify both data and model uncertainties. Physically informed by the Grad–Shafranov equation, EFIT-Prime applies a constraint on the current density J tor and a smoothness constraint on the first derivative of the poloidal flux, ensuring physically plausible solutions. Furthermore, the spatial location of the diagnostics is explicitly incorporated in the inputs to account for their spatial correlation. Extensive evaluations demonstrate EFIT-Prime's accuracy and robustness across diverse scenarios, most notably showing good generalization on negative-triangularity discharges that were excluded from training. Timing studies indicate an ensemble inference time of 15 ms for predicting a new equilibrium, offering the possibility of plasma control in real-time, if the model is optimized for speed.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗