Search NASA⌕ Search

SEARCH · Search NASA

Results for “target configuration”

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
All sourcesNASA NTRSDOE OSTI
NASA NTRS2019

Airborne Radar for sUAS Sense and Avoid

A primary challenge for the safe integration of small UAS operations into the National Airspace System (NAS) is traffic deconfliction, both from manned and unmanned aircraft. The UAS Traffic Management (UTM) project being conducted at the National Aeronautics and Space Administration (NASA) considers a layered approach to separation provision, ranging from segregation of operations through airspace volumes (geofences) to autonomous sense and avoid (SAA) technologies for higher risk, densely occupied airspace. Cooperative SAA systems, such as Automatic Dependent Surveillance-Broadcast (ADS-B) and/or vehicle-to-vehicle communication systems provide significant additional risk mitigation but they fail to adequately mitigate collision risks for non-cooperative (non-transponder equipped) airborne aircraft. The RAAVIN (Radar on Autonomous Aircraft to Verify ICAROUS Navigation) flight test being conducted by NASA and the Mid-Atlantic Aviation Partnership (MAAP) was designed to investigate the applicability and performance of a prototype, commercially available sUAS radar to detect and track non-cooperative airborne traffic, both manned and unmanned. The radar selected for this research was a Frequency Modulated Continuous Wave (FMCW) radar with 120 degree azimuth and 80 degree elevation field of view operating at 24.55GHz center frequency with a 200 MHz bandwidth. The radar transmits 2 watts of power thru a Metamaterial Electronically Scanning Array antenna in horizontal polarization. When the radar is transmitting, personnel must be at least 1 meter away from the active array to limit nonionizing radiation exposure. The radar physical dimensions are 18.7cm by 12.1cm by 4.1cm and it weighs less than 820 grams making it well suited for installation on small UASs. The onboard, SAA capability, known as ICAROUS, (Independent Configurable Architecture for Reliable Operations of Unmanned Systems), developed by NASA to support sUAS operations, will provide autonomous guidance using the traffic radar tracks from the onboard radar. The RAAVIN set of studies will be conducted in three phases. The first phase included outdoor, ground-based radar evaluations performed at the Virginia Tech’s Kentland Farm testing range in Blacksburg, VA. The test was designed to measure how well the radar could detect and track a small UAS flying in the radar’s field of view. The radar was used to monitor 5 test flights consisting of outbound, inbound and crossing routes at different ranges and altitudes. The UAS flown during the ground test was the Inspire 2, a quad copter weighing less than 4250 grams (10 pounds) at maximum payload. The radar was set up to scan and track targets over its full azimuthal field of view from 0 to 40 degrees in elevation. The radar was configured to eliminate tracks generated from any targets located beyond 2000 meters from the radar and moving at velocities under 1.45 meters per second. For subsequent phases of the study the radar will be integrated with a sUAS platform to evaluate its performance in flight for SAA applications ranging from sUAS to manned GA aircraft detections and tracking. Preliminary data analysis from the first outdoor ground tests showed the radar performed well at tracking the vehicle as it flew outbound and repeatedly maintained a track out to 1000 meters (maximum 1387 meters) until the vehicle slowed to a stop to reverse direction to fly inbound. As the Inspire flew inbound tracks from beyond 800 meters, a reacquisition time delay was consistently observed between when the Inspire exceeds a speed of 1.45 meters per second and when the radar indicated an inbound target was present and maintained its track. The time delay varied between 6 seconds to over 37 seconds for the inbound flights examined, and typically resulted in about a 200 meter closure distance before the Inspire track was maintained. The radar performed well at both acquiring and tracking the vehicle as it flew crossing routes out past 400 meters across the azimuthal field of view. The radar and ICAROUS software will be integrated and flown on a BFD-1400-SE8-E UAS during the next phase of the RAAVIN project. The main goal at the conclusion of this effort is to determine if this radar technology can reliably support minimum requirements for SAA applications of sUAS. In particular, the study will measure the range of vehicle detections, lateral and vertical angular errors, false and missed/late detections, and estimated distance at closest point of approach after an avoidance maneuver is executed. This last metric is directly impacted by sensor performance and indicates its suitability for the task.

Szatkowski, George N.↗
DOE OSTI2020

Neutronics analyses for the conceptual design of the SNS Second Target Station

The Spallation Neutron Source, in operation at the Oak Ridge National Laboratory since 2006, was designed to allow the addition of a second target station and an upgrade of the accelerator proton power. Both upgrades are now underway. Here we describe the evolution of the design of the target of the second target station with the emphasis on the effects of the proton beam footprint on the energy deposition in the target, stresses induced by the pulsed operation, and the importance of the residual heat. The moderator configurations and their optimization are discussed. With the utilization of pure parahydrogen moderators, small neutron beam cross-sections, and the specific optimization, neutron beams of the second target station will achieve exceptionally high peak brightness and time-averaged brightness.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2022

Relevance of E × B drifts for particle and heat transport in divertors

Radial electric fields up to ~4 kV m –1 are observed in the boundary between the private flux region (PFR) and the scrape-off layer (SOL) driving E × B drifts between the inner and outer targets at speeds up to 2.8 km s –1 in the Tokamak à configuration variable divertor. The resulting E × B fluxes, located in a narrow region ($\Delta {\rho _\Psi } < 0.012$ in normalized radius or $\Delta $ R – R sep <4 mm mapped to the outer midplane) are equivalent to around 20% of the total heat and particle flux to the divertor targets (inner + outer). At the peak E r , the E × B poloidal transport is equivalent to parallel flows with M ∥ ~ 3. In the snowflake divertor with a second X-point in the outer SOL, the drifts in the PFR-SOL boundary were equivalent to around 30% of the total heat and particle flux to the divertor targets and cover a region ~50% wider than in the single null ($\Delta {\rho _\Psi }$ ~ 0.018, $\Delta $R – R sep ~ 6 mm). Furthermore, the location of the PFR-SOL boundary drift shifts radially in the E ∥ × B direction when reversing the toroidal field direction. Peaks in density and electron pressure have been identified near the primary X-point along with large gradients in density, temperature, and potential, the latter resulting in a local electric field ~2.7 kV m –1 which drives a drift (1.9 km s –1 ) upwards towards the closed flux surfaces. Floating potential (V f ) magnitudes up to 75 V (~2 kT e ) were measured, indicating that V f and parallel currents should not be neglected when estimating plasma potential.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2022

Gluon dipole factorisation for diffractive dijets

Within the colour dipole picture for deep inelastic scattering at small Bjorken x , we study the production of a pair of relatively hard jets via coherent diffraction. By “relatively hard” we mean that the transverse momenta of the two jets — the quark ($q$) and the antiquark ($\overline{q}$) generated by the decay of the virtual photon — are much larger than the target saturation momentum Q s (Y P ) evaluated at the rapidity gap Y P . We argue that the typical final-state configurations are such that the hard $q$$\overline{q}$ dijets are accompanied by a semi-hard gluon jet, with a transverse momentum of the order of Q s (Y P ). The presence of this third jet ensures that the scattering is strong and thus avoids the strong suppression of exclusive (hard) dijet production due to colour transparency. For such “2+1” jet configurations, we demonstrate that both the emission of the semi-hard gluon and its scattering with the hadronic target can be factorised in terms of an effective gluon-gluon dipole. This effective description, originally proposed in [1–7], builds a bridge between the colour dipole picture and the transverse-momentum dependent (TMD) version of the collinear factorisation: the cross-section for diffractive 2+1 jets can be written as the product between a hard factor describing the $q$$\overline{q}$ dijets and a semi-hard factor expressing the unintegrated gluon distribution of the Pomeron. The latter is controlled by gluon dipole scattering in the black disk limit and hence is strongly sensitive to gluon saturation. By integrating out the kinematics of the 3 jets, we obtain the $q$$\overline{q}$$g$ contribution to the diffractive structure function in collinearly-factorised form.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Performance and Durability of Hybrid Fuel Cell Systems for Class-8 Long Haul Trucks

Hybrid fuel cell-battery configurations are investigated that overcome thermal management issues in fuel cell powertrains for heavy-duty Class 8 trucks. The battery is sized so that it has sufficient capacity to provide supplemental power and energy on a hill climb transient at end-of-life. A dynamic load sharing strategy is developed to distribute the power demand between the fuel cell system (FCS) and the energy storage system in a manner that optimizes their lifetimes. The FCS end-of-life is identified as the terminal point beyond which the stack cannot generate the rated power with target power density at 0.7 V and 40 °C ambient temperature. Reaching the target lifetime with a-Pt/C cathode catalyst in one hybrid configuration requires voltage clipping to 813 mV, idle power limited to 50 kW, catalyst overloading to 0.45 mg cm -2 total Pt in anode and cathode, and 44% active membrane area oversizing. The stack and FCS drive cycle efficiencies decrease by 4.2% and 5.4%, respectively, during the electrode lifetime. Further, the FCS performance, durability and cost are compared with the targets of 68% peak efficiency, 0.30 mg cm -2 total Pt loading, 2.5 kW/g PGM Pt group metal (PGM) loading, 750 mW cm -2 power density, 25,000-h lifetime and $80/kW cost.

33 ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2021

Grid Cyber-Security Strategy in an Attacker-Defender Model

The progression of cyber-attacks on the cyber-physical system is analyzed by the Probabilistic, Learning Attacker, and Dynamic Defender (PLADD) model. Although our research does apply to all cyber-physical systems, we focus on power grid infrastructure. The PLADD model evaluates the effectiveness of moving target defense (MTD) techniques. We consider the power grid attack scenarios in the AND configurations and OR configurations. In addition, we consider, for the first time ever, power grid attack scenarios involving both AND configurations and OR configurations simultaneously. Cyber-security managers can use the strategy introduced in this manuscript to optimize their defense strategies. Specifically, our research provides insight into when to reset access controls (such as passwords, internet protocol addresses, and session keys), to minimize the probability of a successful attack. Our mathematical proof for the OR configuration of multiple PLADD games shows that it is best if all access controls are reset simultaneously. For the AND configuration, our mathematical proof shows that it is best (in terms of minimizing the attacker's average probability of success) that the resets are equally spaced apart. We introduce a novel concept called hierarchical parallel PLADD system to cover additional attack scenarios that require combinations of AND and OR configurations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Energy Efficient, Cost-Effective Power and Co-Generation Technologies: Techno-Environmental Analysis

Development of energy solutions for addressing grid resiliency and energy efficiency while lowering greenhouse gas emissions is critical in today’s energy scenario. Chemical energy provides on demand power. Cogeneration technologies offer numerous benefits in meeting the growing energy demand while lowering the impact on environment. Utilization of waste heat from prime movers in conjunction with energy efficient heat pumps and renewable photovoltaics is an attractive approach. Efficient utilization of available resources to support current and future building energy needs targeting grid resiliency, energy and environmental security via co-generation approaches is the focus of this study. A detailed techno-environmental analysis of hybrid system configurations consisting of conventional and emerging technologies utilizing natural gas, electric grid, and renewable power resources along with heat recovery systems and heat pump technologies are analyzed and presented. The key objective is to present integrated system configurations and thermodynamic analysis of various co-generation systems suitable for providing building energy. Design solutions targeting low carbon footprint and high energy efficiency are presented.

Cheekatamarla, Praveen↗
DOE OSTI2026

IER-620 CED-3b: Experiment Execution Summary for the Pulsed-Neutron Die-Away Experiments (PNDA) with Propylene Glycol and Mobilmet 423

There is a strong need for new benchmarks to validate neutron thermal scattering laws (TSLs). Lawrence Livermore National Laboratory (LLNL) has designed a Pulsed-Neutron Die Away (PNDA) testbed for this purpose. The experiment has a deuterium-tritium (D-T) neutron generator that impinges a 10 -4 s, mono-energetic pulse of 14.1 MeV neutrons on a target sample. After the pulse, the neutron population moderates and establishes a thermal equilibrium within the sample, with a fundamental spatial mode and characteristic decay-time eigenvalue, ⍺. The ⍺ eigenvalue can be extracted from the experimental measurements of the time-dependent neutron flux coming off the surface of the sample and can then be used as an integral parameter (similar to k eff ) to validate nuclear data involved with neutron migration, thermalization, and absorption. For moderating materials and geometric configurations, the ⍺ eigenvalue is heavily dependent on thermal neutron scattering of the target material. For that reason, a PNDA experiment can have a higher sensitivity to TSLs than is commonly available with the k eff parameter in critical experiments.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
NASA NTRS1988

Sonic boom loudness study and airplane configuration development

Sonic boom wave form parameters as related to loudness were investigated analytically. The parameters studied include rise time, duration, maximum overpressure and initial overpressure. The design criteria of a 72 dBA for corridors and 65 dBA for unconstrained flight were chosen based on a review of human response testing. The 72 dBA criterion suggests that 1.0 psf shock waves may be acceptable. On that basis, acceptable low sonic boom wave forms were explored with respect to cruise conditions, aerodynamic lifting length requirements and configuration design at M 1.5 and M 2.4. An M 2.4 baseline arrow wing configuration was studied as a possible vehicle for M 1.5 cruise overland. Modifications made to approach the low boom wave form included a slightly longer forebody, staggered nacelles, a lifting arrow wing horizontal tail, and carefully tailored lift and volume elements. The same wave form criteria applied for M 2.4 cruise results in a low boom configuration that has significant weight, length and balance penalties. Further detailed design work is required to reach the target wave form and resultant loudness level for overland cruise at M 1.5. These results so far suggest that a properly designed M 2.4 overwater configuration may be capable of M 1.5 overland operation with sonic boom noise characteristcs that meet the criterion.

Brown, Jessica G.↗
DOE OSTI2023

Post-Shot Report for OMEGA Double Cylinders (CylDRT 22B)

The direct-drive double cylinder experimental platform is a high-energy-density (HED) science platform designed to image an imploding cylindrical target. The target consists of a directly-driven outer cylinder and a shock-and-collision driven inner cylinder. The purpose of this platform is to study hydrodynamic instability growth on the inner cylinder, the outer surface of which is classically Rayleigh-Taylor unstable during the acceleration phase. We present results from recent experiments at the OMEGA laser facility. These experiments were designed as a proof-of-principle for the platform, using the same cylinder exterior dimensions and direct-drive beam configuration as previous single cylinder experiments. In these experiments, three sets of targets were fielded: no machined perturbations (smooth), a sinusoidal mode-10 perturbation on the outer surface of the inner cylinder, and a mode-20 perturbation on the outer surface of the inner cylinder. The primary diagnostic was a gated x-ray framing camera which imaged the backlit inner cylinder on-axis for sixteen frames over a time window of 1 ns for each shot. A second side-lit radiograph captured one image per shot, diagnosing axial uniformity. In this report we include an overview of the results from both the backlighter and the sidelighter diagnostics. We discuss at length the experimental analysis process. Finally, we present the results of the smooth target implosion trajectory and compare them to post-shot simulations. We see favorable agreement between simulation and experiment.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS2009

Mitigation of EMU Glove Cut Hazard by MMOD Impact Craters on Exposed ISS Handrails

Recent cut damages to crewmember extravehicular mobility unit (EMU) gloves during extravehicular activity (EVA) onboard the International Space Station (ISS) has been found to result from contact with sharp edges or pinch points rather than general wear or abrasion. One possible source of cut-hazards are protruding sharp edged crater lips from impact of micrometeoroid and orbital debris (MMOD) particles on external metallic handrails along EVA translation paths. During impact of MMOD particles at hypervelocity an evacuation flow develops behind the shock wave, resulting in the formation of crater lips that can protrude above the target surface. In this study, two methods were evaluated to limit EMU glove cut-hazards due to MMOD impact craters. In the first phase, four flexible overwrap configurations are evaluated: a felt-reusable surface insulation (FRSI), polyurethane polyether foam with beta-cloth cover, double-layer polyurethane polyether foam with beta-cloth cover, and multi-layer beta-cloth with intermediate Dacron netting spacers. These overwraps are suitable for retrofitting ground equipment that has yet to be flown, and are not intended to protect the handrail from impact of MMOD particles, rather to act as a spacer between hazardous impact profiles and crewmember gloves. At the impact conditions considered, all four overwrap configurations evaluated were effective in limiting contact between EMU gloves and impact crater profiles. The multi-layer beta-cloth configuration was the most effective in reducing the height of potentially hazardous profiles in handrail-representative targets. In the second phase of the study, four material alternatives to current aluminum and stainless steel alloys were evaluated: a metal matrix composite, carbon fiber reinforced plastic (CFRP), fiberglass, and a fiber metal laminate. Alternative material handrails are intended to prevent the formation of hazardous damage profiles during MMOD impact and are suitable for flight hardware yet to be constructed. Of the four materials evaluated, only the fiberglass formed a less hazardous damage profile than the baseline metallic target. Although the CFRP laminate did not form any noticeable crater lip, brittle protruding fibers are considered a puncture risk. In parallel with EMU glove redesign efforts, modifications to metallic ISS handrails such as those evaluated in this study provide the means to significantly reduce cut-hazards from MMOD impact craters.

Christiansen, Eric L.↗
DOE OSTI2024

A Planar-Cavity Receiver Configuration for High-Temperature Solar Thermal Processes: Preprint

Next generation concentrating solar thermal power (CSP) and novel solar thermochemical systems using concentrating solar thermal (CST) energy require high operating temperatures exceeding those of traditional nitrate-salt CSP systems. Particle-based systems are attractive for next-generation CSP and CST applications owing to high-temperature stability of inert silica- or alumina-based particulate materials, the lack of low-temperature freezing concerns that limit molten salt and/or molten metal heat transfer media, and cost-effective thermal storage using low-cost particulate and containment materials. Open-cavity falling particle receivers have many potential advantages, but face challenges pertaining to scalability, thermal loss, and particle loss through the open aperture, and are infeasible for chemical processes that require a low-oxygen ambient environment. Enclosed receiver configurations can be scalable, avoid particle loss when heating particles, and have potential for future chemical processes; however, particle-based heat transfer media provide substantially lower heat transfer rates than liquid media, and thus enclosed particle receiver designs require novel configurations to limit surface temperatures under the high incident solar flux concentrations necessary for high receiver thermal efficiency at high temperature. This paper introduces the novel planar-cavity enclosed particle receiver configuration in which arrays of planar surfaces are arranged into sub-vertical cavities. Large angles between the panel surface normal vectors and the aperture surface normal allow the incoming solar beam to distribute along the panel walls. Correspondingly, a high incident solar flux concentration at the cavity aperture produces substantially lower absorbed solar flux concentration on any panel wall. Sets of individual vertical cavities can be arranged to form a scalable receiver configuration.

cavity receiver↗
NASA NTRS1997

Analysis of Technology for Compact Coherent Lidar

In view of the recent advances in the area of solid state and semiconductor lasers has created new possibilities for the development of compact and reliable coherent lidars for a wide range of applications. These applications include: Automated Rendezvous and Capture, wind shear and clear air turbulence detection, aircraft wake vortex detection, and automobile collision avoidance. The work performed by the UAH personnel under this Delivery Order, concentrated on design and analyses of a compact coherent lidar system capable of measuring range and velocity of hard targets, and providing air mass velocity data. The following is the scope of this work. a. Investigate various laser sources and optical signal detection configurations in support of a compact and lightweight coherent laser radar to be developed for precision range and velocity measurements of hard and fuzzy targets. Through interaction with MSFC engineers, the most suitable laser source and signal detection technique that can provide a reliable compact and lightweight laser radar design will be selected. b. Analyze and specify the coherent laser radar system configuration and assist with its optical and electronic design efforts. Develop a system design including its optical layout design. Specify all optical components and provide the general requirements of the electronic subsystems including laser beam modulator and demodulator drivers, detector electronic interface, and the signal processor. c. Perform a thorough performance analysis to predict the system measurement range and accuracy. This analysis will utilize various coherent laser radar sensitivity formulations and different target models.

Amzajerdian, Farzin↗
DOE OSTI2024

Coil optimization for quasi-helically symmetric stellarator configurations

Filament-based coil optimizations are performed for several quasi-helical stellarator configurations, beginning with the one from Landreman & Paul ( Phys. Rev. Lett. , vol. 128, 2022, 035001), demonstrating that precise quasi-helical symmetry can be achieved with realistic coils. Several constraints are placed on the shape and spacing of the coils, such as low curvature and sufficient plasma–coil distance for neutron shielding. The coils resulting from this optimization have a maximum curvature 0.8 times that of the coils of the Helically Symmetric eXperiment (HSX) and a mean squared curvature 0.4 times that of the HSX coils when scaled to the same plasma minor radius. When scaled up to reactor size and magnetic field strength, no fast particle losses were found in the free-boundary configuration when simulating 5000 alpha particles launched at $3.5\,\mathrm {MeV}$ on the flux surface with a normalized toroidal flux of $s=0.5$ . An analysis of the tolerance of the coils to manufacturing errors is performed using a Gaussian process model, and the coils are found to maintain low particle losses for smooth, large-scale errors up to amplitudes of approximately $0.15\,\mathrm {m}$ . Another coil optimization is performed for the Landreman–Paul configuration with the additional constraint that the coils are purely planar. Visual inspection of the Poincaré plot of the resulting magnetic field-lines reveal that the planar modular coils alone do a poor job of reproducing the target equilibrium. Additional non-planar coil optimizations are performed for the quasi-helical configuration with $5\,\%$ volume-averaged plasma beta from Landreman et al. ( Phys. Plasma , vol. 29, issue 8, 2022, 082501), and a similar configuration also optimized to satisfy the Mercier criterion. The finite beta configurations had larger fast-particle losses, with the free-boundary Mercier-optimized configuration performing the worst, losing approximately $5.5\,\%$ of alpha particles launched at $s=0.5$ .

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Information-entropy-driven generation of material-agnostic datasets for machine-learning interatomic potentials

In contrast to their empirical counterparts, machine-learning interatomic potentials (MLIAPs) promise to deliver near-quantum accuracy over broad regions of configuration space. However, due to their generic functional forms and extreme flexibility, they can catastrophically fail to capture the properties of novel, out-of-sample configurations, making the quality of the training set a determining factor, especially when investigating materials under extreme conditions. We propose a novel automated dataset generation method based on the maximization of the information entropy of the feature distribution, aiming at an extremely broad coverage of the configuration space in a way that is agnostic to the properties of specific target materials. The ability of the dataset to capture unique material properties is demonstrated on a range of unary materials, including elements with the FCC (Al), BCC (W), HCP (Be, Re and Os), graphite (C), and trigonal (Sb, Te) ground states. MLIAPs trained to this dataset are shown to be accurate over a range of application-relevant metrics, as well as extremely robust over very broad swaths of configurations space, even without dataset fine-tuning or hyper-parameter optimization, making the approach extremely attractive to rapidly and autonomously develop general-purpose MLIAPs suitable for simulations in extreme conditions.

36 MATERIALS SCIENCE↗
NASA NTRS1989

Calibration of polarimetric radar systems

A practical technique is proposed for calibrating single-antenna polarimetric radar systems using a metal sphere plus any second target with a strong cross-polarized radar cross section. It is shown that all magnitudes and phases (relative to one of the like-polarized linear polarization configurations) of the radar transfer function can be calibrated without knowledge of the scattering matrix of the second target. Comparison of values measured (using this calibration technique) for a tilted cylinder at X-band with theoretical values shows agreement with + or - 0.3 dB in magnitude and + or - 5 deg in phase. The technique is particularly useful for calibrating a radar under field conditions because it does not require careful alignment of calibration targets.

Sarabandi, K.↗
DOE OSTI2024

Experimental capabilities of the LMJ-PETAL facility

In this study, recent progress in the experimental capabilities of the LMJ-PETAL laser facility is reviewed. Updates on the indirect-drive D 2 implosion experiments and equation-of-state experiments using the LMJ laser are presented, including the commissioning of new plasma diagnostics. Several recent campaigns using the PETAL laser alone are also presented, namely the development of a platform using high-resolution and high-energy X-ray sources for radiography experiments, laser wakefield acceleration studies in the self-modulated regime, and neutron generation using a Target Normal Sheath Accelerated proton beam in a pitcher-catcher configuration.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

Uncertainty-driven dynamics for active learning of interatomic potentials

Machine learning (ML) models, if trained to data sets of high-fidelity quantum simulations, produce accurate and efficient interatomic potentials. Active learning (AL) is a powerful tool to iteratively generate diverse data sets. In this approach, the ML model provides an uncertainty estimate along with its prediction for each new atomic configuration. If the uncertainty estimate passes a certain threshold, then the configuration is included in the data set. Here we develop a strategy to more rapidly discover configurations that meaningfully augment the training data set. The approach, uncertainty-driven dynamics for active learning (UDD-AL), modifies the potential energy surface used in molecular dynamics simulations to favor regions of configuration space for which there is large model uncertainty. The performance of UDD-AL is demonstrated for two AL tasks: sampling the conformational space of glycine and sampling the promotion of proton transfer in acetylacetone. The method is shown to efficiently explore the chemically relevant configuration space, which may be inaccessible using regular dynamical sampling at target temperature conditions.

97 MATHEMATICS AND COMPUTING↗