Date of Award
Doctor of Philosophy (PhD)
Dr. Oliver Myers
Dr. Garrett Pataky
Dr. Gang Li
Dr. Zhaoxu Meng
Dr. Asha Hall
The development of composite materials for structural components necessitates methods for evaluating and characterizing their damage states after encountering loading conditions. Laminates fabricated from carbon fiber reinforced polymers (CFRPs) are lightweight alternatives to metallic plates; thus, their usage has increased in performance industries such as aerospace and automotive. Additive manufacturing (AM) has experienced a similar growth as composite material inclusion because of its advantages over traditional manufacturing methods. Fabrication with composite laminates and additive manufacturing, specifically fused filament fabrication (fused deposition modeling), requires material to be placed layer-by-layer. If adjacent plies/layers lose adhesion during fabrication or operational usage, the strength of the material decreases. When adhesion decreases between adjacent plies/layers, voids can form in the interior of a laminate resulting in delamination---localized, weakened regions. Nondestructive evaluation (NDE) is one method for characterizing the damage state of a material without causing alteration.
Previous literature has demonstrated an embedded magnetostrictive particulate (MSP) NDE technique for CFRP composites. This approach is viable for CFRP laminates and specific AM techniques because of the layer-by-layer fabrication methodology. Noted drawbacks of this approach were the sensitivity to pocket with a high concentration of Terfenol-D particles and the inability to ascertain delamination location based solely on the induced sensing voltage plots. This research proposes the development of an integrated smart composite material formed by embedding a thin film sputtered with Terfenol-D inside an AM beam. The Terfenol-D acts as an embedded sensor intended to permit the use of the previously investigated magnetostriction NDE method. The evolution of AM presents scalable production techniques for future structural components and now enables the production of fiber-reinforced composite parts. Thin films of Terfenol-D, deposited via sputtering, provide a uniform distribution of the giant magnetostrictive material on the surface of a substrate. The thickness of the thin films can be controlled to the nanometer. Depositing the thin films onto a thermoplastic plate provides a sensing element for integrating into an AM part and adhering using high temperatures to induce melting and bonding of the thermoplastic with the structural material.
This research aims to utilize machine learning (ML) as a predictive NDE and design tool for structural composite materials. Supervised ML algorithms are capable of identifying abnormalities associated with damage within composite materials. Additional uses of ML seek to improve the characterization of the magnetoelastic response of Tefenol-D using symbolic regression to define potential functions to describe the stress-strain and strain-magnetic field intensity relationship---relationships important for analytically and numerically modeling the magnetostrictive material in NDE applications. The far-reaching goal is to utilize multiple combined tools of embedded sensing with ML for prediction to forward the performance of structural composite systems.
Nelon, Christopher, "Damage Detection with an Integrated Smart Composite using a Magnetostriction-based Nondestructive Evaluation Method: Integrating Machine Learning for Prediction" (2023). All Dissertations. 3538.
Author ORCID Identifier
Artificial Intelligence and Robotics Commons, Databases and Information Systems Commons, Electro-Mechanical Systems Commons, Manufacturing Commons, Numerical Analysis and Scientific Computing Commons, Other Mechanical Engineering Commons