Abstarct: In this thesis, a diglycidyl ether of bisphenol A epoxy (DGEBA) is reinforced by spherical silica nanoparticles of three different diameters of 17 nm, 25 nm and 65 nm up to 6 wt.%. The aim of this study was to investigate the effect of particle size, particle content and combined use of them on Young's modulus, fracture toughness, fracture energy, impact strength and energy absorption capability of thin-walled square columns and frusta. Experimental results showed that the addition of nanoparticles to epoxy resin and also increasing the particle content increased the Young's modulus, fracture toughness, fracture energy and impact resistance so that this increasing was so clear in the hybrid composite. In the hybrid series and in 6 wt% of the nanoparticles, the largest increase compared to pure epoxy was equivalent 13.3% for Young's modulus, 63.1% for fracture toughness, 138.5% for the fracture energy and 20.1% for impact strength. The results revealed that the effect of particle size on mentioned parameters was negligible. With addition of the nanoparticles, the energy absorption characteristics of thin-walled structures suffered and by increasing the particle content, the reduction was intensified. In each experiment the effectiveness mechanisms of nanoparticles has been studied by scanning electron microscopy (SEM) and it was found that the dominant mechanisms were crack path deflection, crack tip pinning and bowing, debonding and plastic void growth. Finally, the experimental results modeled by artificial neural network and response surface methodology (linear and quadratic) and "the selected model" was identified and reported. According to response surface methodology and genetic algorithms, optimal nanoparticle size and content for various parameters have been reported.