Electrochemical Impedance Pattern Recognition for Detection of Hidden Chemical Corrosion on Aircraft Components. Phase 1.

Abstract

This investigation addressed the need for diagnostic instrumentation compatible with performing the Nondestructive Evaluation (NDE) of hidden chemical corrosion with a high degree of accuracy, sensitivity and versatility on both titanium and aluminum alloys currently used in Air Force and commercial aircraft. The overall approach was directed towards development of pattern recognition schemes based upon the on-line data acquisition of Fast Fourier Transform Electrochemical Impedance Spectroscopy (FFTEIS) instrumentation from the suspect hidden chemical corrosion site. Resulting impedance patterns were then analyzed by application of a Neural Network pattern recognition scheme. The Neural Network Analysis (NNA) was then trained to both detect and grade the severity of hidden corrosion present on the aircraft metal substrate interface of interest. Nueral Net Analysis of FFTEIS data was verified as a powerful diagnostic strategy for in situ hidden corrosion process identification, quantitative analysis and severity grading. Correlations between impedance measurements and corrosion depth were verified by subsequent SEM and EDX examination of the metal interfacial regions. The approach will also be powerful for gaining fundamental information into the nature of corrosion processes and conditions leading to their inception at hidden sites. jg

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Feb 08, 1995
Accession Number
ADA291345

Entities

People

  • A. F. Sammells
  • J. S. Bowers

Tags

Communities of Interest

  • Advanced Electronics
  • Air Platforms

DTIC Thesaurus Topics

  • Acquisition
  • Aircraft Equipment
  • Aircrafts
  • Artificial Intelligence
  • Computer Programs
  • Computers
  • Data Acquisition
  • Detection
  • Fast Fourier Transforms
  • Identification
  • Instrumentation
  • Materials
  • Measurement
  • Neural Networks
  • Pattern Recognition
  • Recognition
  • Time Intervals

Readers

  • Neural Network Machine Learning.
  • Surface Coatings Technology.
  • Theoretical Analysis.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks