Development of multiple cycle coupling suspension in complex system optimization

The design of complex engineering systems requires an initial decomposition of the system into subsystems. Couplings link these systems together, by transferring output data from one subsystem to the input of another. Because complex engineering systems can have hundreds or thousands of such couplings, the optimization of these systems is often quite lengthy. To reduce the optimization time, it becomes important that a system designer have the ability to select couplings that have little effect on the solution accuracy, to temporarily remove them. Previous coupling strength analysis methods have not analytically related the effect of a coupling's removal for multiple cycles to solution accuracy. The method presented in this paper locates weak couplings based on their analytical relationship to the objective function and constraints in the overall optimization problem. Discussion of the application of this new method follows, as well as implementation on a decomposed analytical problem. The method significantly reduces the number of subsystem analyses required to optimize the decomposed problem by suspending couplings for multiple design cycles.