Towards Optimal Design of Composite I-Beams

Professor Mark E. Tuttle
Department of Mechanical Engineering
University of Washington
Seattle, Washington

Sponsored by the University Research Council

Date:  Tuesday, November 10, 1998
Time:  3:30 p.m.
Place:  W128 Nebraska Hall


Thin-walled beams are used extensively in innumerable structural applications, including trusses, truck chassis, towers, bridges, and aerospace structures, to name but a few. Advanced polymeric composites are now being considered for use in all of these areas. As is well known, the stiffness and strength of a laminated composite depends wholly on the stacking sequence used. For example, the bending stiffness of a rectangular composite beam can be very low or very high, depending on the number of plies used and the fiber orientation within each ply. During design of a composite beam the engineer must therefore account for the stacking sequence(s) used throughout the beam, as well as the overall dimensions of the beam cross-section. Furthermore, in the case of a composite I-beams consideration must be given to the manufacturing process used to produce the beam, since the manufacturing process may often place restrictions on possible stacking sequences.

In this study a method of predicting various composite beam stiffnesses (e.g., axial, bending, or torsional stiffnesses) is combined with a global optimization algorithm called Improving-Hit-and-Run to identify the "optimal" design of a composite I-beam. The problem is formulated such that the stacking sequence resulting in maximum axial beam stiffness is identified, while still maintaining minimum required values of bending and torsional stiffnesses. Limitations on fiber angles, imposed by the specific manufacturing process used to fabricate the I-beam, are included in the problem formulation.