MULTI-OBJECTIVE OPTIMIZATION OF LAMINATED COMPOSITE PLATE WITH DIFFUSED LAYER ANGLES USING NON-DOMINATED SORTING GENETIC ALGORITHM (NSGA-II)

摘要

An approach to design the laminated composite plate subjected to in-plane loading conditions is investigated: The composite structure is optimized for the conflicting multi-objectives maximum buckling load factor and minimum weight. Fibre orientation angles, stacking sequence and number of layers are chosen as design variables. In order to avoid the inter-laminar stress developed due to grouping of plies. design of laminated plate with diffused stacking sequences is proposed in this work. In-plane strength of the laminate and not grouping more than four plies of same orientation are set as the design constraints. Static failure criteria (Tsai-Wu and Maximum Stress) are used to determine the load bearing capacity for a configuration generated during the optimization process. The effectiveness of the proposed approach is explored with two optimization problems. In the first problem a simple genetic algorithm (SGA) is applied to maximize the buckling load factor of the composite plate. In the second problem, a multi-objective genetic algorithm (NSGA-II) is used to obtain Pareto optimal designs for the problem having multiple, conflicting, objectives minimum weight and maximum buckling load factor. The results show that better design can be obtained with diffused angle layers and the grouping of plies can also be completely eliminated.