Lay-Up Optimization of Composite Stiffened Panels Using Linear Approximations in Lamination Space

摘要

Finding a nearby design with continuous optimization as a first step is a standard approach for reducing the cost of discrete optimization. The authors' previous work on the design of composite stiffened panels used lamination parameters for this first step, and then sought the discrete design that was the closest in lamination parameter space to the continuous design (minimum-distance approach). In this note, a new second-step optimization that used a GA to find the safest design based on a first-order Taylor series about the continuous optimum was developed.rnFor the same thicknesses of the laminates, the new second step generated designs with lower critical constraint violation than those found using the minimum-distance approach. This was used to further reduce the thicknesses of the laminates. It was found that the new second step produced optimum discrete designs with a potential mass saving of 3.5% corresponding to stiffener type a. Furthermore, it was observed that the new optimum discrete designs were further away and in a substantially different direction, and thus were not close in the lamination parameter space to the continuous optima.