Buckling of functionally graded hydrogen-functionalized graphene reinforced beams based on machine learning-assisted micromechanics models

Nanocomposite reinforced with functionalized graphene is a novel class of high-performance materials with great potential in developing advanced lightweight structures in a wide range of engineering applications. However, accurate estimation of its material properties at different temperature conditions remains a great challenge as existing micromechanics models fail to capture the effects of chemical functionalization and temperature. This paper develops machine learning (ML)-assisted micromechanics models by employing genetic programming (GP) algorithm and molecular dynamics (MD) simulation to address this key scientific problem. The well-trained ML-assisted Halpin-Tsai model and rule of mixture can accurately and efficiently predict the temperature-dependent material properties including Young's modulus, Poisson's ratio, coefficient of thermal expansion (CTE), and density of hydrogen-functionalized graphene (HFGr) reinforced copper nanocomposites with high coefficients of determination (R2). Then, the buckling behavior of functionally graded (FG) HFGr nanocomposite beams is studied with the aid of the ML-assisted micromechanics models. A detailed parametric study is performed with a particular focus on the effects of hydrogenation percentage, graphene content, and temperature on the buckling performance of the FG-HFGr beam. Results show that bonding more hydrogen functional groups on the HFGr can effectively improve the buckling resistance of the beam.