Key points are not available for this paper at this time.
Multifunctional composites provide more than one function from the same part. The anisotropy, material, and process characterization challenges and the lack of standardization on the 3D-printed multifunctional carbon composites make it difficult for application into aerospace. The current solutions for additive manufacturing (AM) technologies and additively manufactured monofunctional and multifunctional composites are not mature enough for safety-critical applications. A new approach is proposed to explore the use of machine learning (ML) in the design, development, AM, testing, and certification of multifunctional composites for aircraft, unmanned aircraft systems (UAS), and spacecraft. In this work, an artificial neural network (ANN) architecture is proposed. An AM-embedded building block approach integrates the complete lifecycle of aircraft, UAS, and spacecraft using ANN to support the continued operational safety (COS) of aircraft, spacecraft, and UAS. The proposed method exploits the power of ANN on the metadata for the characterization of multifunctional material properties and processes and the mapping of the failure modes compared with the predicted models and history. This paper provides an in-depth analysis and explanation of the new methods needed to overcome the existing barriers, problems, and situations.
Ghimire et al. (Wed,) studied this question.