MULTI-OBJECTIVE OPTIMAL DESIGN OF HYBRID LAMINATES USING CONTINUOUS ANT COLONY METHOD

Hybrid laminated composites are gaining increasing popularity in applications where improved mechanical properties as well as reduced weight and cost are desired. However, the highly nonlinear nature of the governing equations and the combination of continuous and discrete design variables has rendered the problem a challenging one. This paper implements the nature-inspired method of continuous ant colony optimization (ACO(R)) for hybrid laminates comprising high-stiffness graphite/epoxy and low-stiffness glass/epoxy. The number of surface and core layers and their layouts (discrete) as well as the fiber angles in each layer (continuous) form the design variable set; whereas the composite's fundamental frequency and cost (to be maximized and minimized, respectively) are taken as the objective functions. Comparison of the results with those reported in the literature having used standard discrete fiber angles and implemented methods such as the ant colony system (ACS) and GA, indicates a clear improvement in the values of the objective functions for the majority of the studied cases. It is nonetheless recommended that, for practical purposes, the closest standard fiber angles to the optimal values be adopted (a procedure generally being used but not proven to be optimal), as the corresponding deviations would be negligible.