Mehul Patel

A high-resolution integrated model of the National Ignition Campaign cryogenic layered experiments

A detailed simulation-based model of the June 2011 National Ignition Campaign cryogenic DT experiments is presented. The model is based on integrated hohlraum-capsule simulations that utilize the best available models for the hohlraum wall, ablator, and DT equations of state and opacities. The calculated radiation drive was adjusted by changing the input laser power to match the experimentally measured shock speeds, shock merger times, peak implosion velocity, and bangtime. The crossbeam energy transfer model was tuned to match the measured time-dependent symmetry. Mid-mode mix was included by directly modeling the ablator and ice surface perturbations up to mode 60. Simulated experimental values were extracted from the simulation and compared against the experiment. Although by design the model is able to reproduce the 1D in-flight implosion parameters and low-mode asymmetries, it is not able to accurately predict the measured and inferred stagnation properties and levels of mix. In particular, the measu...

Simulating solidification in metals at high pressure: The drive to petascale computing

We investigate solidification in metal systems ranging in size from 64,000 to 524,288,000 atoms on the IBM BlueGene/L computer at LLNL. Using the newly developed ddcMD code, we achieve performance rates as high as 103 TFlops, with a performance of 101.7 TFlop sustained over a 7 hour run on 131,072 cpus. We demonstrate superb strong and weak scaling. Our calculations are significant as they represent the first atomic-scale model of metal solidification to proceed, without finite size effects, from spontaneous nucleation and growth of solid out of the liquid, through the coalescence phase, and into the onset of coarsening. Thus, our simulations represent the first step towards an atomistic model of nucleation and growth that can directly link atomistic to mesoscopic length scales.

Modification of the Flow Structure over a UAV Wing for Roll Control

Plasma enhanced aerodynamics was used to provide roll control at high angles of attack on a scaled 1303 UAV configuration. The 1303 planform has a 47 degree leading-edge sweep angle. The flow over the a half-span model was documented with dye flow visualization in a water tunnel for a range of angles of attack. This revealed a complex flow structure that varied with angle of attack. A half-span model with Single Dielectric Barrier Discharge (SDBD) plasma actuators was then tested in a wind tunnel. The model was mounted on a 2-D force balance designed to measure lift and drag. At larger angles of attack from 10 to 35 degrees, plasma actuators placed just below the leading edge were found to augment the lift. This configuration was implemented in a full-span model that was mounted on a sting that allowed free-to-roll motion. The ability of the plasma actuator arrangement to produce roll maneuvers was then investigated for a range of angles of attack and freestream speeds. The results indicated excellent roll control with roll moment coefficients that are comparable to conventional moving surfaces.

Performance of indirectly driven capsule implosions on the National Ignition Facility using adiabat-shaping

A series of indirectly driven capsule implosions has been performed on the National Ignition Facility to assess the relative contributions of ablation-front instability growth vs. fuel compression on implosion performance. Laser pulse shapes for both low and high-foot pulses were modified to vary ablation-front growth and fuel adiabat, separately and controllably. Three principal conclusions are drawn from this study: (1) It is shown that reducing ablation-front instability growth in low-foot implosions results in a substantial (3-10X) increase in neutron yield with no loss of fuel compression. (2) It is shown that reducing the fuel adiabat in high-foot implosions results in a significant (36%) increase in fuel compression together with a small (10%) increase in neutron yield. (3) Increased electron preheat at higher laser power in high-foot implosions, however, appears to offset the gain in compression achieved by adiabat-shaping at lower power. These results taken collectively bridge the space between t...

Towards a more universal understanding of radiation drive in gas-filled hohlraums

We have found that radiation-hydrodynamic calculations that use the high flux model assumptions [1] can accurately predict the radiation drive produced by a laser-heated hohlraum under certain conditions, but can not predict drive over a broad range of parameters (pulse energy, hohlraum gas fill density, hohlraum case-to-capsule ratio). In particular, the model is accurate for ~7 ns long laser pulses used to implode capsules with high density carbon (HDC) ablators in hohlraums with helium fill gas densities of 0-0.6 mg/cc. By systematically varying the gas fill density from 0 to 1.6 mg/cc we found that the agreement with drive begins to diverge for fills > 0.85 mg/cc. This divergence from the model coincides with the onset of measureable SRS backscatter. In this same set of experiments the radiation drive symmetry inferred from the imploded shape of a gas-filled capsule is not predicted with this model. Finally, several possible fixes to the model to reduce the observed discrepancies are considered.

Radius Flow Vectoring for Projectile Drag and Steering Control Using Plasma Actuators

This work involves the application of Single Dielectric Barrier Discharge (SDBD) plasma actuators for radius flow vectoring on the blunt aft portion of a generic body of revolution in order to modify the aerodynamic forces. The application is to reduce the pressure drag in the wake and/or produce asymmetric loads that could be used for flight control. The experiments are performed over a range of free-stream Mach numbers from 0.3 to 0.5. The model consisted of a 2.09 cm diameter by 16.0 cm long body of revolution with a elliptic tip. It was mounted on a 5-component load cell so that two force and three moment components could be measured. The experiments only focused on drag and side force. As a pre-cursor, lower-speed measurements on two-dimensional, 90 ◦ trailing edges of different radii were used to examine the performance and scaling of plasma actuators for flow vectoring around a circular radius. This was was documented with particle tracers introduced upstream in the boundary layer, and surface flow visualization. The results indicated that an optimum location for the actuator was slightly upstream of the flow separation location. They indicated a minimum sensitivity to free-stream speed and boundary layer thickness, and a linear dependence of the flow turning angle with plasma actuator voltage. These results were applied to the design of the body of revolution. The baseline drag measurements on the body followed a trend with an average drag coefficient of 0.13. The results with an axisymmetric plasma actuator at two voltages gave an average drag coefficient of 0.09, or approximately 30% lower. There was particular scatter in the results near Mach 0.4 that was thought to be the result of boundary layer transition to turbulence on the model. If these points were dropped, the average drag reduction was a more modest 9%. A similar plasma actuator that covered 180 ◦ of the circumference of the model was used to produce a side force by vectoring the wake. For the same Mach number and actuator conditions of the drag reduction experiments, this produced a mean change in the side force lift coefficient of 17%.

BlueGene/L applications: Parallelism On a Massive Scale

BlueGene/L (BG/L), developed through a partnership between IBM and Lawrence Livermore National Laboratory (LLNL), is currently the worlds largest system both in terms of scale, with 131,072 processors, and absolute performance, with a peak rate of 367 Tflop/s. BG/L has led the last four Top500 lists with a Linpack rate of 280.6 Tflop/s for the full machine installed at LLNL and is expected to remain the fastest computer in the next few editions. However, the real value of a machine such as BG/L derives from the scientific breakthroughs that real applications can produce by successfully using its unprecedented scale and computational power. In this paper, we describe our experiences with eight large scale applications on BG/ L from several application domains, ranging from molecular dynamics to dislocation dynamics and turbulence simulations to searches in semantic graphs. We also discuss the challenges we faced when scaling these codes and present several successful optimization techniques. All applications show excellent scaling behavior, even at very large processor counts, with one code even achieving a sustained performance of more than 100 Tflop/s, clearly demonstrating the real success of the BG/L design.

Testing nonlocal models of electron thermal conduction for magnetic and inertial confinement fusion applications

Three models for nonlocal electron thermal transport are here compared against Vlasov-Fokker-Planck (VFP) codes to assess their accuracy in situations relevant to both inertial fusion hohlraums and tokamak scrape-off layers. The models tested are (i) a moment-based approach using an eigenvector integral closure (EIC) originally developed by Ji, Held, and Sovinec [Phys. Plasmas 16, 022312 (2009)]; (ii) the non-Fourier Landau-fluid (NFLF) model of Dimits, Joseph, and Umansky [Phys. Plasmas 21, 055907 (2014)]; and (iii) Schurtz, Nicolai, and Busquets [Phys. Plasmas 7, 4238 (2000)] multigroup diffusion model (SNB). We find that while the EIC and NFLF models accurately predict the damping rate of a small-amplitude temperature perturbation (within 10% at moderate collisionalities), they overestimate the peak heat flow by as much as 35% and do not predict preheat in the more relevant case where there is a large temperature difference. The SNB model, however, agrees better with VFP results for the latter problem if care is taken with the definition of the mean free path. Additionally, we present for the first time a comparison of the SNB model against a VFP code for a hohlraum-relevant problem with inhomogeneous ionisation and show that the model overestimates the heat flow in the helium gas-fill by a factor of ?2 despite predicting the peak heat flux to within 16%.

Yaw Control of a Blunt-Nose Projectile at High Angles of Attack Using Strakes

An experimental study was conducted to investigate the effects of different aftbody strakes on a projectile with a blunt-nose and a fineness ratio of 4. The effect of strake parameters such as shape, locations (axial and azimuthal), deployment height, and in some cases, the number of strakes implemented was examined. The main objective for the study is to identify promising strake configurations for effective yaw stabilization and control, and to identify changes in the effect of actuator parameters as a function of angle of attack. Wind tunnel experiments were conducted for angles of attack ranging from 0 to 64 deg at a Reynolds number of 0.19 x 10 6 and Mach 0.1. A few test cases were conducted to examine the effect of sideslip angles. The optimum azimuthal location for a strake was found to be the left and right side meridians and 1-inch (x/L = 0.083) from the nose apex. Large yaw control authority was attained for α > 40 deg. The largest yaw control authority was produced by a rectangular-shaped strake. The yaw control attained with this strake was close to symmetric with the strake placed at the corresponding left and right side meridians, and produced a side force and yawing moment to the opposite side of where it was mounted. Aftbody strakes were effective even at sideslip conditions and with larger fins.

3D Simulations of the NIF indirect drive ignition target design

The radiation hydrodynamics code Hydra is used to quantify the sensitivity of different NIF ignition point designs to several 3D effects. Each of the 48 NIF quads is included in the calculations and is allowed to have different power. With this model we studied the effect on imploded core symmetry of discrete laser spots (as opposed to idealized azimuthally-averaged rings) and random variations in laser power.