Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Fusion Science”

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 19 records

Simulation of Intense Beams and Targets for Heavy-Ion-Fusion Science (HEDLP / Inertial Fusion Energy)

Our principal goals, and activities in support of those goals, over the next five years are as follows: (1) Optimize the properties of the NDCX-II beam for each class of target experiments; achieve quantitative agreement with measurements; develop improved machine configurations and operating points. To accomplish these goals, we plan to use Warp to simulate NDCX-II from source to target, in full kinetic detail, including first-principles modeling of beam neutralization by plasma. The output from an ensemble of Warp runs (representing shot-to-shot variations) will be used as input to target simulations using ALE-AMR on NERSC, and other codes. (2) Develop enhanced versions of NDCX-II (the machine is designed to be extensible and reconfigurable), and carry out studies to define a next-step ion beam facility. To accomplish these goals, much of the work will involve iterative optimization employing Warp runs that assume ideal beam neutralization downstream of the accelerator. (3) Carry out detailed target simulations in the Warm Dense Matter regime using the ALE-AMR code, including surface tension effects, liquid-vapor coexistence, and accurate models of both the driving beam and the target geometry. For this we will need to make multiple runs (to capture shot-to-shot variations), and to both develop and employ synthetic diagnostics (to enable comparison with experiments). The new science that will be revealed is the physics of the transition from the liquid to vapor state of a volumetrically superheated material, wherein droplets are formed, and wherein phase transitions, surface tension and hydrodynamics all play significant roles in the dynamics. These simulations will enable calculations of equation of state and other material properties, and will also be of interest for their illumination of the science of droplet formation.

43 PARTICLE ACCELERATORS↗

HEAVY ION FUSION SCIENCE VIRTUAL NATIONAL LABORATORY 1ST QUARTER 2010 MILESTONE REPORT: Simulations of fast correction of chromatic aberrations to establish physics specifications for implementation on NDCX-1 and NDCX-2

This milestone has been accomplished. The Heavy Ion Fusion Science Virtual National Laboratory has completed simulations of a fast correction scheme to compensate for chromatic and time-dependent defocusing effects in the transport of ion beams to the target plane in the NDCX-1 facility. Physics specifications for implementation in NDCX-1 and NDCX-2 have been established. This milestone has been accomplished. The Heavy Ion Fusion Science Virtual National Laboratory has completed simulations of a fast correction scheme to compensate for chromatic and time-dependent defocusing effects in the transport of ion beams to the target plane in the NDCX-1 facility. Physics specifications for implementation in NDCX-1 and NDCX-2 have been established. Focal spot differences at the target plane between the compressed and uncompressed regions of the beam pulse have been modeled and measured on NDCX-1. Time-dependent focusing and energy sweep from the induction bunching module are seen to increase the compressed pulse spot size at the target plane by factors of two or more, with corresponding scaled reduction in the peak intensity and fluence on target. A time-varying beam envelope correction lens has been suggested to remove the time-varying aberration. An Einzel (axisymmetric electric) lens system has been analyzed and optimized for general transport lines, and as a candidate correction element for NDCX-1. Attainable high-voltage holdoff and temporal variations of the lens driving waveform are seen to effect significant changes on the beam envelope angle over the duration of interest, thus confirming the utility of such an element on NDCX-1. Modeling of the beam dynamics in NDCX-1 was performed using a time-dependent (slice) envelope code and with the 3-D, self-consistent, particle-in-cell code WARP. Proof of concept was established with the slice envelope model such that the spread in beam waist positions relative to the target plane can be minimized with a carefully designed Einzel lens waveform and transport line. WARP simulations have verified the efficacy of the Einzel lens while including more detailed beam physics. WARP simulations have also indicated some unpredicted transittime effects, and methods are currently being explored to compensate and reduce this complication. We have explored the use of an Einzel lens, or system of Einzel lenses, to compensate for chromatic aberrations in the beam focal spot in the NDCX-2 target plane. The final beam manipulations in NDCX-2 (linear velocity ramp, charge neutralization, high field final focus solenoid) are similar to NDCX-1 though the NDCX-2 beam has much higher energy and current. The most relevant distinctions are that the pulse duration at the entrance to the drift compression section is tenfold shorter, and that the beam energy tenfold higher, than in NDCX-1. Placing a time-dependent, envelope angle correcting element at the neutralized drift region entrance presents a very significant challenge to voltage holdoff and voltage swing V(t) in a single Einzel lens. Placing the Einzel lens(es) further upstream reduces the required voltage risetime V'(t) to effect the necessary envelope correction, while increasing the duration over which the timedependent voltage must vary. While this simplifies the technological challenge of designing and operating a Einzel lens in NDCX-2, it does require much finer control of the correcting waveform and measurements of its effect on space-charge dominated beams over a much longer axial path length to target than in the NDCX-1.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Visualization and Analysis in Support of Fusion Science (Final Progress Report)

This report summarizes the results of the award for “Visualization and Analysis in Support of Fusion Science.” With this award our main efforts have been to develop and deploy visualization and analysis tools in three areas 1) magnetic field line analysis 2) query based visualization and 3) comparative visualization.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Long Range Community Planning Process for Fusion Sciences and Plasma Physics (Final Technical Report)

This grant facilitated the APS-DPP Community Planning Process (DPP-CPP) which was a strategic planning process with the goals of identifying scientific and technological opportunities in the fields of Plasma Physics and Fusion Energy Science, and making consensus recommendations for a strategy to address these opportunities to the DOE Office of Fusion Energy Science Advisory Committee (FESAC). The DPP-CPP was initiated by the APS-DPP executive committee in order to help FESAC fulfill a charge from DOE to develop a strategic plan for the DOE Office of Fusion Energy Science (FES). In this charge, the DOE Deputy Director for Science, Dr. Stephen Binkley, requests “that the Fusion Energy Sciences Advisory Committee (FESAC) undertake a new long-range strategic planning activity for the Fusion Energy Sciences (FES) program. The strategic planning activity - to encompass the entire FES research portfolio (namely, burning plasma science and discovery plasma science) - should identify and prioritize the research required to advance both the scientific foundation needed to develop a fusion energy source, as well as the broader FES mission to steward plasma science.” The process as set out in the proposal (and below) was successful in meeting all of its stated goals culminating in the final report to the FESAC subcommittee given the charge.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Uncertainty guided online ensemble for non-stationary data streams in fusion science

Machine Learning (ML) is poised to play a pivotal role in the development and operation of next-generation fusion devices. Fusion data shows non-stationary behavior with distribution drifts, resulted by both experimental evolution and machine wear-and-tear. ML models assume stationary distribution and fail to maintain performance when encountered with such non-stationary data streams. Online learning techniques have been leveraged in other domains, however it has been largely unexplored for fusion applications. In this paper, we investigate online learning for continuous adaptation to drifting data streams in the prediction of Toroidal Field (TF) coils deflection at the DIII-D fusion facility. We further address the short-term performance degradation inherent to standard online learning, which arises because ground truth is unavailable at prediction time. To mitigate this issue, we propose an uncertainty-guided online ensemble framework. The method leverages the Deep Gaussian Process Approximation (DGPA) for calibrated uncertainty estimation and uses these uncertainty measures to guide a meta-algorithm that aggregates predictions from learners trained over different historical horizons. Our results show that online learning reduces prediction error by 80% compared to a static model. The online ensemble and the proposed uncertainty-guided ensemble further reduce error by approximately 6%, and 10% respectively, relative to standard single-model online learning, while also providing calibrated uncertainty estimates to support operational decision-making.

AI↗

Maximizing 1D “like” implosion performance for inertial confinement fusion science

While the march towards achieving indirectly driven inertial confinement fusion at the NIF has made great progress, the experiments show that multi-dimensional effects still dominate the implosion performance. Low mode implosion symmetry and hydrodynamic instabilities seed by capsule mounting features appear to be two key limiting factors for implosion performance. One reason these factors have a large impact on the performance of ICF implosions is the high convergence required to achieve high fusion gains. To tackle these problems, a predictable implosion platform is needed meaning experiments must trade-off high gain for performance. To this end, LANL has adopted three main approaches to develop a 1D implosion platform where 1D means high yield over 1D clean calculations. Taking advantage of the properties of beryllium capsules, a high adiabat, low convergence platform is being developed. The higher drive efficiency for beryllium enables larger case-to-capsule ratios to improve symmetry at the expense of drive. Smaller capsules with a high adiabat drive are expected to reduce the convergence and thus increase predictability. The second approach is liquid fuel layers using wetted foam targets. With liquid fuel layers, the initial mass in the hot spot can be controlled via the target fielding temperature which changes the liquid vapor pressure. Varying the initial hot spot mass via the vapor pressure controls the implosion convergence and minimizes the need to vaporize the dense fuel layer during the implosion to achieve ignition relevant hot spot densities. The last method is double shell targets. Unlike hot spot ignition, double shells ignite volumetrically. The inner shell houses the DT fuel and the convergence of this cavity is relatively small compared to hot spot ignition. Radiation trapping and the longer confinement times relax the conditions required to ignite the fuel. Key challenges for double shell targets are coupling the momentum of the outer shell to the inner shell and mixing of the mid-Z material from the inner shell into the fuel. The primary theme for each of these approaches is reduced implosion convergence with the goal of achieving a 1D “like” implosion. Once established, a systematic approach to solving limiting issues for ICF can be undertaken. This presentation will discuss the approaches, results, and plans for each of these campaigns.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Uncertainty based Online Ensemble on Non-Stationary Data for Fusion Science

Machine Learning (ML) is poised to play a pivotal role in the development and operation of next-generation fusion devices. Fusion data shows non-stationary behavior due to drifts in the data. The drifts can arise from both experimental evolution and machine wear-and-tear. ML models assume stationary distribution and fail to maintain performance when encountered with non-stationary data streams.Online learning can be used to continuously adapt the models with new data as it is acquired. However, traditional online learning can suffer from short-term performance degradation, as ground truth are not available before making the prediction. To address this challenge, we propose uncertainty aware ensemble approach for online learning. We use Deep Gaussian Process Approximation (DGPA) technique for calibrated uncertainty estimation and use the uncertainty values to guide a meta-algorithm that produces predictions based on ensemble of learners. Moreover, DGPA also provides uncertainty estimation along with the predictions for decision makers. This paper demonstrates that the proposed method outperforms traditional online learning approach, and a naive ensemble without uncertainty guidance by about 7% and 6%, respectively, on B-coil deflection prediction at DIII-D Fusion Facility.

Rajput, Kishansingh [Thomas Jefferson National Acc↗

Uncertainty based Online Ensemble on Non-Stationary Data for Fusion Science

Machine Learning (ML) is poised to play a pivotal role in the development and operation of next-generation fusion devices. Fusion data shows non-stationary behavior due to drifts in the data. The drifts can arise from both experimental evolution and machine wear-and-tear. ML models assume stationary distribution and fail to maintain performance when encountered with non-stationary data streams.Online learning can be used to continuously adapt the models with new data as it is acquired. However, traditional online learning can suffer from short-term performance degradation, as ground truth are not available before making the prediction. To address this challenge, we propose uncertainty aware ensemble approach for online learning. We use Deep Gaussian Process Approximation (DGPA) technique for calibrated uncertainty estimation and use the uncertainty values to guide a meta-algorithm that produces predictions based on ensemble of learners. Moreover, DGPA also provides uncertainty estimation along with the predictions for decision makers. This paper demonstrates that the proposed method outperforms traditional online learning approach, and a naive ensemble without uncertainty guidance by about 7% and 6%, respectively, on B-coil deflection prediction at DIII-D Fusion Facility.

Rajput, Kishansingh [Thomas Jefferson National Acc↗

Developing one-dimensional implosions for inertial confinement fusion science

Experiments on the National Ignition Facility show that multi-dimensional effects currently dominate the implosion performance. Low mode implosion symmetry and hydrodynamic instabilities seeded by capsule mounting features appear to be two key limiting factors for implosion performance. One reason these factors have a large impact on the performance of inertial confinement fusion implosions is the high convergence required to achieve high fusion gains. To tackle these problems, a predictable implosion platform is needed meaning experiments must trade-off high gain for performance. LANL has adopted three main approaches to develop a one-dimensional (1D) implosion platform where 1D means measured yield over the 1D clean calculation. A high adiabat, low convergence platform is being developed using beryllium capsules enabling larger case-to-capsule ratios to improve symmetry. The second approach is liquid fuel layers using wetted foam targets. With liquid fuel layers, the implosion convergence can be controlled via the initial vapor pressure set by the target fielding temperature. The last method is double shell targets. For double shells, the smaller inner shell houses the DT fuel and the convergence of this cavity is relatively small compared to hot spot ignition. However, double shell targets have a different set of trade-off versus advantages. As a result, details for each of these approaches are described.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Exploration of Quantum Machine Learning and AI Accelerators for Fusion Science

The dawn of the noisy intermediate-scale quantum (NISQ) era sparked rapid development in variational quantum algorithms. These algorithms, utilizing parameterized quantum circuit optimized by classical computers with feedback, are practical under the constraints of the current hardware and can potentially show quantum advantage. In recent years, these variational circuit are applied to neural networks, hoping to boost the triumphant success of deep learning. We simulate various of quantum-classical hybrid neural networks applied to scientific tasks, and seek to understand their capabilities and limitations. Specifically, we devise quantum convolutional layers and apply them to deep neural networks used for predicting plasma disruption in fusion reactors.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

HEAVY ION FUSION SCIENCE VIRTUAL NATIONAL LABORATORY 3nd QUARTER 2009 MILESTONE REPORT: Upgrade plasma source configuration and carry out initial experiments. Characterize improvements in focal spot beam intensity

Simulations suggest that the plasma density must exceed the beam density throughout the drift compression and focusing section in order to inhibit the space charge forces that would limit the spot size and beam intensity on the target. WDM experiments will therefore require plasma densities up to 10{sup 14}/cm{sup 3}, with the highest density in the last few centimeters before the target. This work was guided by the simulations performed for the FY09 Q1 milestone. This milestone has been met and we report results of modifications made to the NDCX beamline to improve the longitudinal and radial distribution of the neutralizing plasma in the region near the target plane. In Section 2, we review pertinent simulation results from the FY09 Q1 milestone. Section 3 describes the design, and beam measurements following installation, of a biased, self-supporting metal grid that produces neutralizing electrons from glancing interception of beam ions. Section 4 describes the design and initial testing of a compact Ferro-Electric Plasma Source (FEPS) that will remove the remaining 'exclusion zone' in the neutralizing plasma close to the target plane. Section 5 describes the modification of the beamline to decrease the gap between the FEPS section exit and the final focus solenoid (FFS). Section 6 presents a summary and conclusions.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Heavy Ion Fusion Science Virtual National Laboratory 4th Quarter 2009 Milestone Report: Measure and simulate target temperature and dynamic response in optimized NDCX-I configurations with initial diagnostics suite

This milestone has been met. The effort contains two main components: (1) Experimental results of warm dense matter target experiments on optimized NDCX-I configurations that include measurements of target temperature and transient target behavior. (2) A theoretical model of the target response to beam heating that includes an equilibrium heating model of the target foil and a model for droplet formation in the target for comparison with experimental results. The experiments on ion-beam target heating use a 300-350-keV K{sup +} pulsed beam from the Neutralized Compression Drift Experiment (NDCX-I) accelerator at LBNL. The NDCX-I accelerator delivers an uncompressed pulse beam of several microseconds with a typical power density of >100 kW/cm{sup 2} over a final focus spot size of about 1 mm. An induction bunching module the NDCX-I compresses a portion of the beam pulse to reach a much higher power density over 2 nanoseconds. Under these conditions the free-standing foil targets are rapidly heated to temperatures to over 4000 K. We model the target thermal dynamics using the equation of heat conduction for the temperature T(x,t) as a function of time (t) and spatial dimension along the beam direction (x). The competing cooling processes release energy from the surface of the foil due to evaporation, radiation, and thermionic (Richardson) emission. A description of the experimental configuration of the target chamber and results from initial beam-target experiments are reported in our FY08 4th Quarter and FY09 2nd Quarter Milestone Reports. The WDM target diagnostics include a high-speed multichannel optical pyrometer, optical streak camera, VISAR, and high-speed gated cameras. The fast optical pyrometer is a unique and significant new diagnostic which provides valuable information on the temperature evolution of the heated target.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗