Abstract
Practical realizations of adaptive structures should be designed and implemented to maximize performance, robustness and efficiency. A method of frequency-shaping the compensation of an adaptive structure is developed for increased performance, system robustness and efficiency. Robustness is defined as reduced coupling to out-of-bandwidth modes, where model fidelity frequently suffers, and to in-bandwidth modes that are considered not important for control. This method utilizes an open-loop metric of system coupling, through Hankel singular values of the controllability and oberservability Gramians, to determine the optimum transducer placements and sizes. The design metric is maximized with genetic algorithms. This optimum design process is computationally efficient. Results demonstrate that optimizing the design metric with genetic algorithms yields better performance and efficiency when compared to previous optimization methods and forms a basic method of “loop-shaping” by spatial design.
Get full access to this article
View all access options for this article.