Skip to main content

Showing 1–3 of 3 results for author: Demkowicz, M J

.
  1. arXiv:2401.12147  [pdf

    math.NA math-ph

    An Efficient Finite Difference-based Implicit Solver for Phase-Field Equations with Spatially and Temporally Varying Parameters

    Authors: Zirui Mao, G. R. Liu, Michael J. Demkowicz

    Abstract: The phase field method is an effective tool for modeling microstructure evolution in materials. Many efficient implicit numerical solvers have been proposed for phase field simulations under uniform and time-invariant model parameters. We use Eyre's theorem to develop an unconditionally stable implicit solver for spatially non-uniform and time-varying model parameters. The accuracy, unconditional… ▽ More

    Submitted 22 January, 2024; originally announced January 2024.

  2. Comparing theoretical predictions of radiation-free velocities of edge dislocations to molecular dynamics simulations

    Authors: Daniel N. Blaschke, Ta Duong, Michael J. Demkowicz

    Abstract: Transonic defect motion is of interest for high strain-rate plastic deformation as well as for crack propagation. Ever since Eshelby's 1949 prediction in the isotropic limit of a 'radiation-free' transonic velocity $v_\text{RF}=\sqrt{2}c_{\textrm{T}}$, where shock waves are absent, there has been speculation about the significance of radiation-free velocities for defect mobility. Here, we argue th… ▽ More

    Submitted 21 November, 2023; v1 submitted 11 May, 2023; originally announced May 2023.

    Comments: 16 pages, 7 figures, 2 tables; v2+v3 expanded revision, v4 title changed

    Report number: LA-UR-23-25095

    Journal ref: Phys. Rev. B 108 (2023) 224102

  3. arXiv:2210.06734  [pdf, other

    eess.SY

    Optimal Control of Material Micro-Structures

    Authors: Aayushman Sharma, Zirui Mao, Haiying Yang, Suman Chakravorty, Michael J Demkowicz, Dileep Kalathil

    Abstract: In this paper, we consider the optimal control of material micro-structures. Such material micro-structures are modeled by the so-called phase field model. We study the underlying physical structure of the model and propose a data based approach for its optimal control, along with a comparison to the control using a state of the art Reinforcement Learning (RL) algorithm. Simulation results show th… ▽ More

    Submitted 13 October, 2022; originally announced October 2022.