Descriptor-Guided Screening of Band-Selective Elastic-Wave Anisotropy in Reaction–Diffusion Phononic Media

Published: 29 May 2026| Version 1 | DOI: 10.17632/tstsf4tpsh.1
Contributors:
Chung Il Park,
,
,
,

Description

This dataset contains processed simulation outputs and compact MATLAB scripts supporting the manuscript “Descriptor-Guided Screening of Band-Selective Elastic-Wave Anisotropy in Reaction–Diffusion Phononic Media.” The dataset includes 49 two-dimensional Gray–Scott reaction–diffusion morphologies generated over a 7×7 (F,k) parameter grid, processed morphology descriptors, mapped two-phase elastic-property data, and precomputed scalar and shear-horizontal anisotropy metrics. The deposited descriptors include dominant wavelength, interface density, ring-filtered spectral anisotropy, nematic orientational order, connectivity-related measures, labyrinth index, and maturity score. The anisotropy data correspond to the fair band-selection procedure used in the manuscript, where the frequency-ordered branch and normalized wave-number radius giving the largest angular frequency anisotropy are selected for each morphology. The accompanying MATLAB scripts load the processed dataset, validate the 7×7 case structure, regenerate descriptor/anisotropy tables, and reproduce the main manuscript-style Figures 1–4 from the processed data. The scripts are intended for transparent reproduction of the deposited analysis outputs; they do not rerun the full Gray–Scott time integration or the full plane-wave eigensolver calculations.

Files

Steps to reproduce

Unpack the dataset folder and open MATLAB from the package root directory. Then run: addpath(genpath('code')) run_reproduce_main_figures The script validates the 7×7 processed dataset, regenerates descriptor/anisotropy tables, and writes reproduced versions of the main manuscript Figures 1–4 to the reproduced_figures/ directory. The scripts use deposited processed data and do not rerun the Gray–Scott time integration or plane-wave eigensolvers.

Institutions

Categories

Engineering, Computational Mechanics

Funders

Licence