LAUSR

It is well-known that polymer nanocomposites can bring about superior mechanical, thermal, optical, physical, and chemical properties in comparison with pure polymers. In this study, different contents of unmodified silica nanoparticles (Si-Un), surface modified nano-silica by octylsilane (Si-OS), and surface modified nano-silica by polydimethylsiloxane (Si-PDMS) are added to the polyurethane (PU) matrix and their effects on the physical properties of the polymer examined. The experimental results indicate that most of the nanocomposites have a higher tensile strength and elongation. In addition, hyperelastic energy function models have been used to model the stress-strain relation of the nanocomposites. In this study, Mooney-Rivlin, neo-Hookean, Rivlin general polynomial, and Davies-De Thomas (DDT) models have been investigated, possessing respectively, two, one, eight, and three constants to be determined. The differential evolution (DE) optimization method, a strong heuristic optimization algorithm, has been used to find the constants; in which the absolute summation of the differences between the models’ predictions and experimental data is taken into account as the objective function and the models’ constants are considered as the decision variables. Moreover, equation constants are found by using regression, an indicator of DE optimization superiority. The results show that even though the Rivlin general polynomial model provides the most accurate prediction, the DDT model, consisting of three constants, can be considered as the most acceptable one.