Modeling the mechanical properties of Composites Reinforced by Silica nanoparticles through Response Surface Method & Regression Tree Method

Ali Dadrasi¹ Abdolreza Alavi Gharebagh² Sasan Fooladpanjeh³

1) Technical Engineering Faculty of Islamic Azad University, Shahrood Branch
2) Computer and Electrical Engineering Faculty of Islamic Azad University, Shahrood Branch
3) Master’s students in Islamic Azad University, Shahrood Branch

Publication : 3rd international conference of Science and Engineering(3icesconf.com)
Abstract :
An experimental study has been conducted on fracture toughness (KIC) and fracture energy (GIC) of composites reinforced by silica nanoparticles in this article. Three different diameters of 23, 74, and 170 nanometer nanoparticles were added to epoxy resin by up to 30 volume percent. Two parameters, the size of the particles and the volume percent, have been considered as the parameters affecting the above- mentioned properties. Moreover, these two parameters have been considered as the input parameters for the purpose of modeling the results through the response surface model method & regression tree method. The experimental results and the results of modeling indicate that adding silica nanoparticles has a significant effect on Young s modulus, fracture toughness and fracture energy in such a manner that these parameters increase when the silica nanoparticle content is added however the size of the particles does not have a significant effect on them. Also the experimental results and the modeling results show that the second order response surface model makes the best predictions. Additionally, the best value for Young s modulus is 5.801 GPa when the particle size is 170 nm and volume percent is 30. The best value for fracture toughness is equal to 2.852 , when the particle size is 112 nm and the volume percent is 30. Also this method shows that the best value for fracture energy is equal to 1270 J/m2 when the size of the particle is 112 nanometers and the volume percent is 30. Also, compare the result obtained from Regression tree method and RSM method show that still the results obtained from RSM method is much more suitable from Regression Tree method.
Keywords : Nano composite Silica Fracture toughness Response surface method