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

This program addresses the need for diagnostic instrumentation to detect the presence of hidden chemical corrosion occurring at aircraft titanium and aluminum alloys. The approach will be directed towards development of pattern recognition schemes based upon the initial on-line acquisition of electrochemical impedance spectra using Fast Fourier Transform Electrochemical Impedance Spectroscopy (FFTEIS) instrumentation from the suspect corrosion site. This will be followed by implementation of a Neural Network Analysis (NNA) to the impedance data obtained. The resulting NNA transformed data will then be utilized by software based pattern recognition techniques to classify the hidden chemical corrosion site. The experimental program will be performed by: i) fabrication of laboratory FFTEIS instrumentation incorporating preferred components, ii) performing impedance measurements on relevant metal substrates subjected to hidden chemical corrosion, iii) developing and training Neural Networks for unambiguously detecting and determining the severity of hidden chemical corrosion on the subject metal substrates and iv) fabricating and performance testing prototype instrumentation.