فهرست نشریات علمی دسترسی آزاد ایران

Optimizing the mechanical properties of multiscale fiber-reinforced composites, which incorporate nanoparticles for enhanced performance, is a complex challenge due to the non-linear interactions between multiple material parameters. Traditional one-factor-at-a-time experimentation is inefficient, while standalone statistical or heuristic optimization methods often lack either precision or robustness. This study develops a novel hybrid optimization framework that synergistically integrates the Taguchi method for global design-of-experiments screening with Particle Swarm Optimization (PSO) for precise local refinement. The methodology was applied to maximize the tensile strength and elastic modulus of an epoxy composite reinforced with hybrid carbon-Kevlar fibers and hybrid nanoparticles (nano-silica and nano-graphene). Experiments identified an optimal combination of 1.2 wt% nano-silica and 0.75 wt% nano-graphene, achieving a tensile strength of 390 MPa and an elastic modulus of 45 GPa—representing improvements of 85% and 28.5%, respectively, over the non-reinforced baseline. The hybrid Taguchi-PSO framework then refined this result, predicting a precise optimum at 1.18 wt% and 0.78 wt% for the respective nanoparticles, forecasting a tensile strength of 393 MPa (an 86.3% improvement) and elastic modulus of 45.1 GPa (an 28.7% improvement). Critically, the hybrid approach demonstrated superior convergence efficiency and consistency compared to either method used in isolation. The core novelty lies in the intelligent, two-stage sequence where Taguchi efficiently identifies a high-performance region, which then strategically guides and constrains the PSO search, overcoming the individual limitations of each method and providing a robust, systematic tool for the advanced design of multifunctional nanocomposites.