A multispectral nondestructive approach for imagebased damage monitoring

Damage in composite materials even in well-controlled laboratory conditions is a stochastic process. The main reason for the variability in observed damage patterns is primarily the spatial randomness of damage initiation sites, which is a direct result of the current manufacturing procedures for common composite materials. In an effort to provide a practical method for visualizing damage progression that could assist mechanical and damage behavior characterization, the combined use of two emerging non-contact and full field nondestructive methods is presented in this article. Digital Image Correlation (DIC) and passive Infrared Thermography (IR) can be readily applied during testing of composite materials. The advantage of both techniques is that they provide high resolution deformation and thermal fields, respectively. Such fields especially under varying fatigue loading conditions have the potential to provide correlations between macroscopically observed changes in the mechanical behavior and spatially resolved "hot spots", directly related to final failure. Going beyond the simple combined use of the two techniques and based on their individual properties, a path to stitch information extracted from both and perform image-based monitoring of evolving damage in composites is described. Custom data acquisition, noise reduction and post-processing algorithms were created to allow the calculation of features that can provide information related to volume-dependent development of surface damage patterns and their correlations with both mechanical and other nondestructive information (i.e. provided by acoustics). This approach has the potential to provide a quantitative tool for the multispectral description of continuously evolving material states, which if reliably identified can also serve as direct inputs to physics-based computational models for progressing damage quantification and remaining useful life prediction.