Optimization of composite laminate by using genetic algorithm with recessive gene like repair strategy

Abstract

A genetic algorithm is used to obtain the stacking sequence of the laminate that has the set of lamination parameters that are the closest to a set of target lamination parameters. In addition, the laminate is required to be balanced and to have no more than 4 contiguous plies of the same orientation in order to satisfy practical considerations. This problem is a constrained combinatorial optimization problem which is usually difficult to solve. The difficulty of enforcing constraints in genetic optimization is handled by introducing a new repair strategy. The new repair strategy does not alter genes but only changes decoding rules, and is similar in this respect to the way recessive genes operate in biology. The relationship between the reliability of the genetic algorithm and the probability of repair was investigated, and it is shown that 100% probability of repair is optimal when the target laminates include 45° plies. Since practical composite laminates usually include 45° plies, it is concluded that the repair strategy discussed herein should always be used with the optimization.