Title :
Model validation and parameter identification for systems in H∞ and l1
Author_Institution :
Electrical & Computer Eng. Dept., University of California, Santa Barbara, CA 93106, USA
Abstract :
Robust control models contain unknown perturbations with assumed bounds. Modeling of systems in a robust framework requires a means of estimating such bounds. One can begin to address this problem by studying model validation - is there a perturbation and an unknown disturbance, fitting the assumptions of the model, that accounts for the experimental observation? This paper discusses this problem for systems modeled as members of H∞ or l1. The assumptions of these design frameworks are significantly different, giving very different implications for the model validation problem. Model validation can also give an approach to parameter identification in robust control models. Again the consequences of the usual assumptions of each framework are discussed.
Keywords :
Additive noise; Design methodology; Pain; Parameter estimation; Performance analysis; Robust control; Robust stability; Robustness; Strontium; Transfer functions;
Conference_Titel :
American Control Conference, 1992
Conference_Location :
Chicago, IL, USA
Print_ISBN :
0-7803-0210-9