Accurate finite element model of equiaxed-grain engineering material for ultrasonic inspection

Bo Xiao, Richard O'Leary, Anthony Gachagan, Wenqi Li, Timothy Burnett

Research output: Contribution to conferencePaper

166 Downloads (Pure)

Abstract

Finite element (FE) simulation of grained material with equiaxed grain distribution is of interest for the virtual prototyping of array structures and the assessment of signal processing algorithms. Construction of such models can be computationally intensive due to the large number of crystallographic orientations required to represent the material. This paper concentrates on analysis and processing of orientation data in order to establish a computationally efficient 2D FE model whilst maintaining appropriate accuracy of the grained structure. Two approaches for orientation processing are proposed and their performances are compared. Parametric studies show that the trade-off between computational overhead and model accuracy will reach the optimal point when Euler space is segmented with a bin size of 15 degree per Euler phase. A transducer array is then incorporated into the FE model to generate B-scan image of the material. The image is compared with experimental equivalent for FE model validation purpose. The minor difference of images proves that the constructed FE model is accurate, highlighting the potential of the proposed methods for application on other equiaxed-grain materials.
Original languageEnglish
Publication statusPublished - 21 Oct 2015
EventIEEE 2014 International Ultrasonics Symposium - Hilton, Chicago, United Kingdom
Duration: 3 Sept 20146 Sept 2014

Conference

ConferenceIEEE 2014 International Ultrasonics Symposium
Country/TerritoryUnited Kingdom
CityChicago
Period3/09/146/09/14

Keywords

  • material characterization
  • equiaxed-grain material
  • accurate modelling
  • finite element method

Fingerprint

Dive into the research topics of 'Accurate finite element model of equiaxed-grain engineering material for ultrasonic inspection'. Together they form a unique fingerprint.

Cite this