Oxford Centre for High Energy Density Science (OxCHEDS)

Abstract:

We present a crystal plasticity model tailored for high-pressure, high-strain-rate conditions that uses a multiscale treatment of dislocation-based slip kinetics. We use this model to analyze the pronounced plasticity-induced lattice rotations observed in shock-compressed polycrystalline tantalum via in situ x-ray diffraction. By making direct comparisons between experimentally measured and simulated texture evolution, we can explain how the details of the underlying slip kinetics control the degree of lattice rotation that ensues. Specifically, we show that only the highly nonlinear kinetics caused by dislocation nucleation can explain the magnitude of the rotation observed under shock compression. We demonstrate a good fit between our crystal plasticity model and x-ray diffraction data and exploit the data to quantify the dislocation nucleation rates that are otherwise poorly constrained by experiment in the dynamic compression regime.

Abstract:

We present measurements of the temporal decay rate of one-dimensional (1D), linear Langmuir waves excited by an ultrashort laser pulse. Langmuir waves with relative amplitudes of approximately 6% were driven by 1.7J, 50 fs laser pulses in hydrogen and deuterium plasmas of density ne0 = 8.4 × 1017 cm−3. The wakefield lifetimes were measured to be τH2wf = (9 ± 2) ps and τ D2wf = (16 ± 8) ps, respectively, for hydrogen and deuterium. The experimental results were found to be in good agreement with 2D particle-in-cell simulations. In addition to being of fundamental interest, these results are particularly relevant to the development of laser wakefield accelerators and wakefield acceleration schemes using multiple pulses, such as multipulse laser wakefield accelerators.

Abstract:

Optical imaging systems are inherently limited in their resolution due to the point spread function (PSF), which applies a static, yet spatially-varying, convolution to the image. This degradation can be addressed via Convolutional Neural Networks (CNNs), particularly through deblurring techniques. However, current solutions face certain limitations in efficiently computing spatially-varying convolutions. In this paper we propose CoordGate, a novel lightweight module that uses a multiplicative gate and a coordinate encoding network to enable efficient computation of spatially-varying convolutions in CNNs. CoordGate allows for selective amplification or attenuation of filters based on their spatial position, effectively acting like a locally connected neural network. The effectiveness of the CoordGate solution is demonstrated within the context of U-Nets and applied to the challenging problem of image deblurring. The experimental results show that CoordGate outperforms existing approaches, offering a more robust and spatially aware solution for CNNs in various computer vision applications.