Parameter Optimization Algorithm of a Discrete Energy-Averaged Model for Galfenol Alloys

Ismail Nas, Zhangxian Deng, Suryarghya Chakrabarti, Marcelo Dapino

Research output: Contribution to conferencePresentation

Abstract

Magnetostrictive materials deform when exposed to magnetic fields and undergo change in magnetization when stressed. Magnetostrictive iron-gallium alloys, also known as Galfenol, possess a unique combination of mechanical robustness and moderate magnetostriction. Galfenol can withstand bending, tensile and torsional loads and can be machined, welded, or extruded into complex geometries. Thus, it opens up avenues for three-dimensional (3D) functional, structural, and versatile magnetostrictive devices including energy harvesters, sensors, actuators, and mechanical dampers. Fully-coupled 3D modeling frameworks for magnetostrictive materials have been developed in previous studies following an energy-based approach. Taking advantage of cubic symmetric Galfenol, the discrete energy-averaged (DEA) model significantly improves model efficiency by reducing the possible magnetic moment orientations. In this study, a new numerical approximation approach for partial derivative expressions is developed, which improves computational speed of the DEA model by 61% relative to existing partial derivative expressions. A parameter optimization algorithm is proposed to determine the parameters of the discrete energy-averaged (DEA) model for Galfenol alloys. Initial estimation of model parameters and a two-step optimization procedure, including anhysteresis and hysteresis steps, are performed to improve accuracy and efficiency of the algorithm. The performance, contribution, and discretionary usage of each step of the optimization algorithm are evaluated in terms of efficiency and accuracy. Initial estimation of certain material properties such as saturation magnetization, saturation magnetostriction, Young's modulus, and anisotropy energies can improve the convergence and enhance the efficiency by 41% compared to the case where these parameters are not estimated. The two-step optimization improves efficiency by 28% while preserving accuracy compared to one-step optimization. The proposed optimization algorithm is then applied to Galfenol data obtained from the literature. These measurements account for characterization of several samples with different crystal structures, compositions, and heat treatments. The resulting trends can be used to identify model parameters in cases when measurements are not available.
Original languageAmerican English
StatePublished - Sep 2018
Externally publishedYes
EventASME 2017 SMASIS Conference on Smart Materials, Adaptive Structures and Intelligent Systems - Snowbird, UT
Duration: 1 Sep 2018 → …

Conference

ConferenceASME 2017 SMASIS Conference on Smart Materials, Adaptive Structures and Intelligent Systems
Period1/09/18 → …

EGS Disciplines

  • Biomedical Engineering and Bioengineering
  • Mechanical Engineering

Fingerprint

Dive into the research topics of 'Parameter Optimization Algorithm of a Discrete Energy-Averaged Model for Galfenol Alloys'. Together they form a unique fingerprint.

Cite this