Theses and Dissertations
Mechanical and thermal behavior of multiscale bi-nano-composites using experiments and machine learning predictions
Mississippi State University
Thompson, David S.
Newman, James C. Jr.
Date of Degree
Dissertation - Open Access
Doctor of Philosophy
James Worth Bagley College of Engineering
Department of Aerospace Engineering
The mechanical and thermal properties of natural short latania fiber (SLF)-reinforced poly(propylene)/ethylene-propylene-diene-monomer (SLF/PP/EPDM) bio-composites reinforced with nano-clays (NCs), pistachio shell powders (PSPs), and/or date seed particles (DSPs) were studied using experiments and machine learning (ML) predictions. This dissertation embraces three related investigations: (1) an assessment of maleated polypropylene (MAPP) coupling agent on mechanical and thermal behavior of SLF/PP/EPDM composites, (2) heat deflection temperature (HDT) of bio-nano-composites using experiments and ML predictions, and (3) fracture toughness ML predictions of short fiber, nano- and micro-particle reinforced composites. The first project (Chapter 2) investigates the influence of MAPP on tensile, bending, Charpy impact and HDT of SLF/PP/EPDM composites containing various SLF contents. The second project (Chapter 3) introduces two new bio-powderditives (DSP and PSP) and characterizes the HDT of PP/EPDM composites using experiments and K-Nearest Neighbor Regressor (KNNR) ML predictions. The composites contain various contents of SLF (0, 5, 10, 20, and 30wt%), NCs (0, 1, 3, 5wt%), micro-sized PSPs (0, 1, 3, 5wt%) and micro-sized DSPs (0, 1, 3, 5wt%). The third project (Chapter 4) characterizes the fracture toughness of the same composite series used in the second project, by applying Charpy impact tests, finite element analysis, and a ML approach using the Decision Tree Regressor (DTR) and Adaptive Boosting Regressor (ABR). 2wt% MAPP addition enhanced the composite tensile/flexural moduli and strength up to 9% compared with the composites with zero MAPP. In addition, energy impact absorption was profoundly increased (up to78%) and HDT (up to 4 Co) was improved upon MAPP addition to the composites. SLF, NC, DSP and PSP could separately and conjointly increase HDT and fracture toughness values. The KNNR ML approach could accurately predict the composite’s HDT values and, Decision Tree Regressor (DTR) and Adaptive Boosting Regressor ML algorithms worked well with fracture toughness predictions. Pictures taken through a transmission electron microscope, scanning electron microscope and X-Ray proved the NC dispersion and exfoliation as one of the factors in HDT and fracture toughness improvements.
Daghigh, Vahid, "Mechanical and thermal behavior of multiscale bi-nano-composites using experiments and machine learning predictions" (2020). Theses and Dissertations. 3045.
Bio-nano-composites||Machine learning||Natural fibers||Fracture toughness||Heat deflection temperature