Optimum design of multiple configuration structures for frequency constraints

Abstract

Introduction D the service of an aerospace structure, its may vary for different operating conditions. When we restrict the word configuration to structural geometry, the structures with multiple configurations that come to mind are, for example, a tilt-rotor aircraft in airplane or helicopter mode and a stowed vs deployed solar array system for a spacecraft. If we generalize the word to encompass boundary conditions and mass distribution, then all aerospace vehicles can be considered to be structures with multiple configurations. In application, design specifications may impose specific natural frequency constraints for different configurations. For a structure with a single configuration, the minimum weight design with frequency constraints can be found readily.' However, the resulting optimum design may not satisfy frequency constraints for the structure in other configurations. This often leads to design iterations among various configurations. After many design iterations, the final design may be feasible but not optimum. The purpose of this paper is to develop an approach for minimum weight design of structures with several configurations under natural frequency constraints. In the optimum design problem, frequency constraints for all configurations are considered simultaneously. In each design cycle, the constrained minimum weight design problem is solved iteratively by a combined analysis/optimization procedure. To increase the computational efficiency, the iterative analysis is approximated by using reduced-order modal space models.' Sequential linear programming (SLP)'' for optimization procedure is considered. In SLP, the optimum design problem is linearized at each design iteration. Together with a strategy to compute move limits, the resulting linear optimization problem is solved as a linear programming problem. The combination of finite element analysis together with optimization method (SLP) constitutes an effective and reliable approach for solving practical optimum design problems.