Features and defects characterisation for virtual verification and certification of composites: A review
Vincent K. Maes, Kevin Potter, James Kratz
Abstract
Composite manufacturing is driven by a balance between costs (i.e. material and time) and quality. Due to the brittle nature of composite materials, even small deviations in the parts (i.e. defects) can result in significant reductions in load carrying ability of a part. The occurrence of defects is a complex problem, with many sources and factors which affect them. To assist in better understanding and predicting part quality, statistical tools and advanced machine learning can be used to help fill the gaps. A solid understanding of part quality can then in turn be used in combination with a digital twin to achieve virtual testing and certifcation of a part while requiring less physical tests. However, as this review shows, the available data in the literature does not sufficiently characterise key defects, nor their dependence on part design and process parameters to achieve this goal. As such it is argued here that enhanced characterisation and manufacturing trials of more complex parts are needed to generate the required database.