Date of Award

Summer 2024

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Mechanical & Aerospace Engineering

Program/Concentration

Mechanical Engineering

Committee Director

Oleksandr G. Kravchenko

Committee Member

Miltos Kotinis

Committee Member

Gene Hou

Committee Member

Jiang Li

Committee Member

Sergey G. Kravchenko

Abstract

Prepreg platelet molded composites (PPMC) are long, discontinuous fiber reinforced polymer materials. PPMC are an important subcategory of composite materials as they are processible into geometrically complex structures and can be produced via high-throughput manufacturing processes, however they have higher stiffness and strength as compared to traditional discontinuous fiber reinforced polymers. However, there is inherent randomness in the structure of PPMCs and as such, PPMC parts frequently require per part testing that is cost prohibitive.

Herein, a method using artificial intelligence is proposed as a more cost-effective method of inspecting PPMC parts. Different artificial intelligence (AI) architectures are explored to solve this problem. In order to train AI models, a dataset is developed from high fidelity finite element models (FEMs) of PPMC plates. In each FEM a PPMC with a unique local material orientation morphology undergoes a temperature change of 150 °C. The thermally induced strains are extracted from the top and bottom surfaces of the plate, as is local fiber orientation distribution (FOD) data.

Each AI model explored here-in uses the strain fields as an input and outputs PPMC local FOD fields. With the large datasets developed, each AI model is trained to minimize prediction error of local FOD. AI models are first deployed to predict the thickness-average FOD and do so satisfactorily. Next, AI models are developed for prediction of FOD at sub-thickness levels, and finally, an AI model is developed to predict the material orientation at the relevant mesostructural fidelity for full structural reconstruction of the PPMC plate.

The mesostructure reconstruction via artificial intelligence technique is performed for an example 3D part, as well as for the digital twin of a specimen that was non-destructively scanned and translated to a FEM mesh. Further experimental and in-silico validation of the methodology are explored.

Rights

In Copyright. URI: http://rightsstatements.org/vocab/InC/1.0/ This Item is protected by copyright and/or related rights. You are free to use this Item in any way that is permitted by the copyright and related rights legislation that applies to your use. For other uses you need to obtain permission from the rights-holder(s).

DOI

10.25777/hn97-c932

ISBN

9798384456490

ORCID

0000-0001-8525-3557

Download

Share

COinS