A Novel Automated Assessment Approach for Diagnosis of Aircraft Composite Materials Based on Machine Learning Thermographic Images

Abstract

Inspecting, diagnosing, maintaining and predicting aircraft safety faults are among the most essential regular jobs in complex, safety-critical airframes. Moreover, the development of advanced imaging diagnostic tools such as Non-Destructive Testing techniques (NDT), in particular, for aircraft composite materials, has been considered the subject of intense research over the past decades. The need for prompt and reliable diagnostic tools for composite materials in aircraft applications is growing and attracting increasing interest. However, there is still an ongoing need to develop new tools and approaches to respond to the rapid industrial development and complex machine design. These tools will facilitate early detection and isolation of developing defects and prediction of damage propagation. This allows for early implementation of preventative maintenance and acts as a countermeasure to the possibility of catastrophic failure. In this study, following a short introductory summary and definitions, this research presents a brief review of the recent research literature on failure diagnosis of composite materials, and focuses on developing an automated assessment approach using machine learning tools for aerospace composites. However, to date this investigation is unique and offers a significant contribution to the existing body of knowledge on the use of thermography techniques.