Physically based dispersion curve feature analysis in the NDE of composites

Krishnan Balasubramaniam, Joseph Lawrence Rose

Research output: Contribution to journalArticle

23 Scopus citations

Abstract

The oblique incidence of longitudinal waves onto a structure generates different wave modes, each one having unique physical characteristics and hence a different degree of sensitivity to the state of the material. The primary focus, thus far, has been to generate acoustical models and to confirm their integrity. In this paper, an attempt is being made to use the mechanics of plate waves with specific emphasis on solving the anomaly estimation problem for composite materials and to device inspection guidelines using a feature based analysis. Theoretical models for anomalies such as porosity, fiber volume fraction, fiber orientation defects, and hybrid ply layup effects, are all built into an effective material stiffness constants model. The results are then coupled to a generalized plate wave dispersion algorithm. Thus material properties with different anomaly content could be generated and their corresponding plate wave dispersion diagrams computed. Several new features from these dispersion curves were defined and shown to correlate quantitatively with anomaly content. The analytical studies were further supplemented with strong experimental observations, thus defining several new viable features. The promising features could improve anomaly presence prediction because they show an increased the sensitivity to subtle variations in material degradation.

Original languageEnglish (US)
Pages (from-to)41-67
Number of pages27
JournalResearch in Nondestructive Evaluation
Volume3
Issue number1
DOIs
StatePublished - Mar 1 1991

All Science Journal Classification (ASJC) codes

  • Materials Science(all)
  • Condensed Matter Physics
  • Mechanics of Materials
  • Mechanical Engineering

Fingerprint Dive into the research topics of 'Physically based dispersion curve feature analysis in the NDE of composites'. Together they form a unique fingerprint.

Cite this