A Basic Two-Stage Decomposition Strategy for Design Optimization

Abstract

A basic two-stage decomposition strategy for design optimization is described. Resolution of the independent variable vector into two parts is accompanied by a decomposition of the optimization problem equation system into a modular component and an integrating system component. The modular component must be explicitly solvable by an optimization technique, and specifically the method of optimal design is used. The integrating system component is generally the basis of control for an automated optimal design search process of reduced dimensionality. A technique is developed for handling possibilities of incompatible specifications in the modular component. Three examples illustrate application of the developed algorithm.